inert gases in the control of museum insect pests (1998)

Charles SelwitzShin Maekawa

research in

conservatio

n

Inert Gases in theControl of MuseumInsect Pests

The Getty Conservation Institute

The Getty Conservation Institute


1998

research in

conservatio

nCharles SelwitzShin Maekawa

Tevvy Ball, Managing Editor

Elizabeth Maggio, Copy Editor

Amita Molloy, Production Coordinator

Garland Kirkpatrick, Series Designer

© 1998 The J. Paul Getty Trust

All rights reserved.

Printed in the United States of America

10 9 8 7 6 5 4 3 2 1

Library of Congress Cataloging-in-Publication Data

Selwitz, Charles, 1927-

Inert gases in the control of museum insect pests / Charles

Selwitz, Shin Maekawa.p. cm. — (Research in conservation)

Includes bibliographical references (p. ) and index.

ISBN 0-89236-502-1 (pbk.)1. Museum conservation methods—Research. 2. Museum exhibits—

Conservation and restoration. 3. Insect pests—Control—Research.

4. Gases, Rare. 5. Nitrogen. 6. Oxygen. 7. Carbon dioxide.

8. Anoxemia. I. Maekawa, Shin, 1952- . II. Getty Conservation

Institute. III. Title. IV. Series.

AM145.S45 1998069’.53—DC21 98-7310

CIP

The GettyConservationInstitute

Research inConservation

The Getty Conservation Institute works internationally to further the apprecia-tion and preservation of the world's cultural heritage for the enrichment and useof present and future generations. The Institute is an operating program of theJ. Paul Getty Trust.

The Research in Conservation reference series presents the findings of researchconducted by the Getty Conservation Institute and its individual and insti-tutional research partners, as well as state-of-the-art reviews of conservationliterature. Each volume covers a topic of current interest to conservators andconservation scientists. Other volumes in the Research in Conservation seriesinclude Oxygen-Free Museum Cases (Maekawa 1998); Stone Conservation:An Overview of Current Research (Price 1996); Accelerated Aging: Photo-chemical and Thermal Aspects (Feller 1994); Airborne Particles in Museums(Nazaroff, Ligocki, et al. 1993); Epoxy Resins in Stone Consolidation (Selwitz1992); Evaluation of Cellulose Ethers for Conservation (Feller and Wilt 1990);Cellulose Nitrate in Conservation (Selwitz 1988); and Statistical Analysis inArt Conservation Research (Reedy and Reedy 1988).

Contents ix ForewordMiguel Angel Corzo

xi Preface

xv Acknowledgments

1 Chapter 1 Mechanisms of Insect MortalityDesiccationAnoxia at High Humidities

7 Chapter 2 Anoxia as a Conservation ProcedureFrom Food Preservation to ConservationEarly Museum Anoxia StudiesDetermining Kill TimesThermal TechniquesKilling Burrowed Insects

17 Chapter 3 Methods and MaterialsBarrier FilmsGas SuppliesOxygen ScavengersClosures

Heat SealersZippersClampsAdhesive Strips

Portals

29 Chapter 4 Operational Problems and PracticesHumidificationArgon versus NitrogenMonitors

Electronic Oxygen MonitorsPassive Oxygen MonitorsCarbon Dioxide MonitorsRelative Humidity Monitors

Temperature SensorsLeak DetectorsMonitoring Life SignsSafe Use of Nontoxic Fumigants

41 Chapter 5 Anoxia Treatment in Barrier-Film BagsGeneral ProceduresConservators' ExperienceComments

49 Chapter 6 Anoxia Treatment in a Dynamic ModeTreatment in Small PouchesTreatment in Constructed Containments

Objects at the J. Paul Getty MuseumBack Seat Dodge '38The Spanish Piano

vii

57 Chapter 7 Reusable Anoxia SystemsFlexible Chambers

The Getty Barrier-Film TentThe Small Rentokil BubbleThe Large Rentokil Bubble

Rigid ChambersHomemade ContainersVacufume Chambers

67 Chapter 8 Fumigation with Carbon DioxideEffectiveness of Carbon Dioxide versus NitrogenCarbon Dioxide Fumigation RequirementsTreatment Experience in the United States

Old Sturbridge VillageThe Winterthur MuseumSociety for the Preservation of New England AntiquitiesThe Oakland MuseumThe Smithsonian Institution

Treatment Experience in CanadaThe National Museum of Science and TechnologyThe Canadian Museum of Civilization

Treatment Experience in Germany

79 Appendix A: Professional Contacts

81 Appendix B: Materials and Suppliers

91 Appendix C: Common and Scientific Names of Insect Species

93 References

101 Index

107 About the Authors

viii

Foreword The interest of the Getty Conservation Institute in the management of pestcontrol reached a turning point in 1987 when Nieves Valentín, then with theInstituto de Conservación y Restauración de Bienes Culturales (now called theInstituto del Patrimonio Histórico Español) in Madrid, Spain, came to the GCI fora tenure as a Fellow to continue her conservation research on ways to stop bio-logical attack on organic materials. Concerned about the impact that traditionalpesticides have on both conservators' safety and the environment, the Instituterecognized the need to contribute to the development of newer, safer tech-niques for halting the damage done by insects, fungi, and algae to museum col-lections. With this goal in mind, the GCI invited Dr. Valentín to join its staff andpursue her research in this important area of conservation. Her presence at theInstitute and the enthusiasm she generated in her field of endeavor openedfurther avenues of exploration.

Since then, the GCI has continued research in this field both in-house and withassociated institutions. This has resulted in a number of completed projects suchas storage cases for the Royal Mummies of Egypt's National Museum; displaycases to protect the Constitution of India developed in collaboration with theNational Physical Laboratories of that country; cases for mummies in the MuseoBalaguer in Barcelona, Spain; the disinfestation of Kienholz's Back Seat Dodge'38, done in collaboration with the J. Paul Getty Museum and the Los AngelesCounty Museum of Art; the display case for the Hudson Bay Charter, in Toronto;and technical advice for the conservation of the Declaration of Arbroath inScotland. Additionally, the Institute has offered pest control managementcourses in various locations around the world.

Inert Cases in the Control of Museum Insect Pests is a compendium of informa-tion on the biological mechanisms by which nontoxic gases kill insects; themethods and materials needed to create and maintain an anoxic atmosphere;treatments; the construction and use of chambers and bubbles; and the proce-dures for treating objects.

This publication is a continuation of the Institute's Research in Conservationseries, which aims to make available the findings of research conducted by theInstitute and its partners as well as provide state-of-the-art reviews of conserva-tion literature. From its inception the Institute has sought to open to the conser-vation professions the results of its research and experience in areas that bridgethe gap between contemporary science and technology and current conserva-tion practice.

In addition to Nieves Valentín, GCI senior scientist Shin Maekawa and GCIFellows Kerstin Elert and Vinod Daniel have dedicated much of their time tothis area, and a number of other institutions also have collaborated in ourresearch efforts.

With his usual scientific rigor, organizational skills, and attention to detail,Charles Selwitz undertook the exacting task of reviewing the literature and theresearch done at the Institute and by other individuals and organizations to pro-duce this publication. To him, to my colleagues, and to our partners, I extend mysincere acknowledgment.

I hope our readers will find in these pages the information and expertise that willprovide them with further means to continue in our joint efforts to help conservethe world's cultural heritage.

—Miguel Angel CorzoDirectorThe Getty Conservation Institute

ix

Preface Over the centuries, one of the most persistent causes of loss of cultural propertyand museum collections has been damage done by insects. Prime targets aremanuscripts on paper and parchment, natural history collections, and herbaria,but massive wooden objects also are often attacked and, occasionally, seriouslyharmed. Throughout the nineteenth and most of the twentieth centuries, theresponse to this biological assault has been to use an array of highly toxic chemi-cals, an approach that has resulted in many well-documented negative conse-quences. More recently, the conservation community has begun to appreciatethat prevention is a better approach and that there are physical and biochemicalprocedures that can safely substitute for pesticides when prevention is occasion-ally inadequate. As embodied in a program of regular inspections, good house-keeping, and maintenance, prevention is designed to avoid providing pestswith access to collections and is more commonly called integrated pest man-agement. Better and safer killing methods to treat an established infestationinclude the use of biologically active agents such as pyrethrums and juvenilegrowth hormones, and physical means such as desiccation with nontoxic gases,freezing, and heating. Increasingly, well-established pest control programs atboth small and large institutions are being built on the considered use of all ofthese options.

This book is about one option, the use of nontoxic gases (argon, nitrogen, andcarbon dioxide) to rid museum objects of insect pests. Conservators are selectingthis approach because they feel more comfortable and confident that using inertgases is much less likely to harm objects than other procedures. Curators havebeen known to insist that disinfestation be carried out by anoxia (which isdefined as a deficiency of oxygen reaching bodily tissues of such severity as tocause permanent damage) in clear plastic bags so they can keep watch over theircharges during treatment. Freezing and thermal methods are also effective whencarried out properly. They are popular because they allow large quantities ofmaterial to be conveniently transported and treated. But freezing and heating,by their nature, must bring objects to unaccustomed temperatures whereunwanted changes often occur. In contrast, treatment of cultural property inan inert atmosphere rather than air provides a more stable environment wheredeterioration is less likely. For example, the use of a low-oxygen environmentwill deter biological growth, prevent surface oxidation, and retard color fading.Studies by Burke (1992), Arney, Jacobs, and Newman (1979), and others showthat the longevity of most organic colorants is substantially increased in a nitro-gen atmosphere with less than 1000 ppm oxygen. One such study examined anumber of inorganic pigments under these conditions and found that three ofthem—litharge (PbO), cinnabar (HgS), and sienna (mostly Fe2O3)—showedslight color changes after a one-month exposure. Additionally, Valentín (1990)found that low oxygen levels can be used to inhibit the growth of both bacteriaand fungi.

In 1996 the remaining toxic fumigants generally approved for residential usein the United States were Vikane, sulfuryl fluoride, and methyl bromide. Inmuseum facilities, however, the use of toxic gases has been abandoned, pri-marily for reasons of safety. But if it were necessary to find an additional reasonfor choosing nitrogen over Vikane, one might consider the visual effect of thetwo gases on painting materials. Koestler and coworkers (1993) assessed theimpact of the two gases on thirty combinations of linen, rabbit-skin-glue size,lead-white oil ground, and oil-based paints employing eleven different inorganicpigments. The comparison was based on a visual evaluation made by two paint-ings conservators of color change, gloss change, blanching, topography change,and precipitation. Samples treated with Vikane by a commercial exterminatorunderwent the conventional treatment of two, well-spaced short exposures;anoxia samples were kept under nitrogen containing less than 1000 ppm oxygenfor five months. Impurities in the Vikane (hydrogen chloride and sulfur dioxide)adversely affected ten of the eleven pigment samples, while nitrogen had no

xi

observable effect on any sample. The inorganic colorants included lead oxideand sienna, but not cinnabar. These studies demonstrate that although thereare many ways to rid museum collections of insect pests, none is likely to besafer or more protective of the integrity of objects than the use of a controlledatmosphere.

The preceding discussion helps to explain the rapid growth of anoxia and carbondioxide fumigation as conservation procedures. In 1990 these methods were vir-tually unknown in the museum community, although studies were under way ina number of facilities. In most locations these studies quickly gave way to prac-tice. Appreciation of this new preservation technology has since spread rapidlythrough journal publications, papers presented at meetings, and anoxia trainingcourses. Nieves Valentín, of the Instituto del Patrimonio Histórico Español (for-merly the Instituto de Conservación y Restauración de Bienes Culturales) inMadrid, conducted workshops in Spain and Latin America and brought thisapproach to pest control to Spanish institutions such as the Prado Museum, theFundació Joan Miró, the Museum of Decorative Arts, and the National Palace ofFine Arts, as well as the National General Archives in Columbia and the NationalPalace of Fine Arts in Cuba. John Burke, with the Oakland Museum Conserva-tion Center, provided training to conservators in California through a variety offormats, including a workshop at the San Diego Natural History Museum inMarch 1996. And the International Institute for Conservation—Canadian Groupheld a preconference course in May 1996 in Montreal entitled "Those Pests inCollections: Control of Insects and Fungi in Cultural Collections," which featuredcarbon dioxide fumigation. The Getty Conservation Institute offered programson pest management and control for museums in Los Angeles in 1994 and inLondon in 1996.

Another important factor in the dissemination of this technology is the activesharing of new information by experienced conservators and conservation scien-tists, a practice not seen in most other disciplines, and the generous response torequests for help by the early workers in this field. Almost all of these people arestill willing to answer questions and have permitted us to list their names in theProfessional Contacts section of this book (Appendix A). Up to 1997 it waspossible to track the institutions that were using nontoxic gases for pest con-trol, but this is no longer possible because of the rapid growth in this field. Itseems that most cultural institutions, large and small, in Europe, Latin America,Australia, and North America are now considering or actually employing thistechnology, while an increasing number of commercial vendors are making avail-able materials and systems designed specifically for use with nontoxic gases.

The growing interest in anoxia and carbon dioxide fumigation has led to a largenumber of inquiries by museum personnel who are aware of this technology, andit is for this reason that this book was prepared. It is directed and dedicated toconservators who must handle problems of insect infestation, and it is intendedto be a guidebook for this purpose. We attempt to do this by explaining how theproper use of these gases brings about 100% mortality of all life stages of thesmall number of insect species that are problems; by describing how to createanoxia systems in conservation facilities and providing examples of the waymany prominent institutions are using this technology; by listing where suppliesand operating systems can be purchased; and even by providing information onwhom to telephone for help.

Chapter 1 describes the biological mechanisms by which nontoxic gases killinsects. The biochemistry of mortality by nitrogen and argon differs somewhatfrom that of carbon dioxide, but all of these gases owe their effectiveness todesiccation. It is important for conservators to understand this because factorsthat impact on desiccation, such as temperature and humidity, are important inthe design and operation of treatment systems.

xii

The second chapter discusses the pioneering research of conservation scientistsMark Gilberg, with the National Center for Preservation Technology and Trainingin Natchitoches, Louisiana, and Nieves Valentín. Working from different back-grounds, they demonstrated the absolute effectiveness of anoxia as a preserva-tion tool and created practical systems for dealing with insect problems. Thischapter also describes the early research by Michael Rust and Janice Kennedy ofthe University of California, Riverside. They conducted experiments with all lifestages of twelve of the most destructive species, involving tens of thousands ofspecimens, to determine the minimum time necessary to ensure 100% mortalityof each species under nitrogen containing less than 1000 ppm oxygen. Thechapter further examines how factors such as oxygen concentration, type of gas(argon or nitrogen), temperature, relative humidity (RH), the embedment ofinsects inside objects, and the duration of treatment affect the success of the process.

Chapter 3 provides information about the methods and materials needed to cre-ate and maintain an anoxic atmosphere. In all cases, containment must be essen-tially free of oxygen; thus containers must be constructed of materials throughwhich this gas cannot pass. This requires a nitrogen supply with only a few partsper million of oxygen, and container walls or plastic bags made of materials,called barrier films, that have very low oxygen permeability. This chapter alsodescribes the use of powerful oxygen scavengers that can convert air to anacceptable grade of nitrogen. Three types of treatments have evolved from themanner in which these tools are used: maintaining a continuous flow of nitrogenover objects; sealing small items inside plastic pouches holding oxygen scaven-gers; and treating large quantities of material inside reusable chambers.

Chapter 4 provides guidance on aspects of carrying out treatments. The well-being of objects likely to suffer infestation generally depends on maintainingadequate humidity. Humidification is needed for continuous flow treatment,while the addition of water may not be necessary with static systems if theobjects and their packaging provide adequate moisture buffering. Proceduresfor humidification are described. With anoxia systems the operator has a choicebetween nitrogen and argon, and the factors influencing that choice are dis-cussed. Insect mortality is produced 30% faster with argon than with nitrogen.However, nitrogen is generally cheaper than argon, and most conservators haveopted for this gas. Monitoring systems for oxygen and carbon dioxide concen-trations, temperature, and RH are described in a comprehensive section bycoauthor Shin Maekawa. Requirements for the safe use of nitrogen, argon, andcarbon dioxide are also discussed. These gases are not toxic in the conventionalsense, but they must be handled with adequate caution and monitoring. Nitro-gen and argon are suffocants, and carbon dioxide will cause respiratory prob-lems at very modest concentrations.

Chapter 5 provides details on the construction and use of plastic bags made ofoxygen-barrier films. Suitable bags are commercially available, even for large,bulky projects. Rentokil, Inc., for example, has constructed barrier-film bags of80 m3 capacity for treating large numbers of oil paintings with infested woodframes. However, conservators doing a treatment generally make the pouchesby hand in a wide range of sizes and shapes. The application of these bags toproblems at institutions as diverse as the Metropolitan Museum of Art in NewYork, the Mendocino County Museum in Willits, California, the Los AngelesCounty Museum of Art, the Phoebe Apperson Hearst Museum of Anthropologyin Berkeley, and the Oakland Museum is described. Treatments are occasionallycarried out by maintaining a continuous flow of humidified, high-purity nitrogenabout infested material for two or more weeks. This can be done by running thenitrogen into an artifact-containing bag that has a small, deliberate leak. Moreoften this treatment is used with large, awkwardly shaped, and difficult-to-moveobjects. In such cases a complex barrier-film container is built around the object

xiii

with heat-sealed plastic panels. Chapter 6 describes the use of this procedure todisinfest ornate furniture from the decorative arts collection of the J. Paul GettyMuseum, modern sculpture in the form of Back Seat Dodge '38 by EdwardKienholz at the Los Angeles County Museum of Art, and a grand and regalBartolomé March piano in Spain.

Many institutions conduct anoxia treatments in reusable systems that generallyhave a much larger capacity than do pouches. These fall into two classes: flexiblebubbles and rigid chambers. Chapter 7 provides directions for making a 10 m3

barrier-film bubble, but most operators prefer to buy one of the commercialRentokil units. Rigid chambers are perhaps more easily constructed in-housethan are bubbles, and two examples are provided. Using a long, 18 in.(45.72 cm) diameter section of standard poly(vinyl chloride) pipe, Steven Pineof the Museum of Fine Arts in Houston built a chamber to treat items of somelength but modest width, like rolls of fabric. Alan Johnston put together a simplebut effective chamber for the Hampshire Council Museums using a polypropyl-ene shell, a plate of polycarbonate, a large gasket, and several C-clamps.Another option is to locate a commercial fumigation chamber left over fromthe days when treatment with ethylene oxide was in vogue. The Los AngelesCounty Museum of Art has such a unit, which was easily converted for use withhumidified nitrogen and is now run on a routine basis.

The final chapter of this book, chapter 8, deals with the use of carbon dioxidefor treating infested objects. Although anoxia has found more widespread use,the amount of cultural property treated with carbon dioxide is larger. This gasrequires less stringent control of concentration; it is cheaper to use; and the acid-ity created by carbonic acid formation has not been found harmful to treatedobjects. Although the biological mechanisms leading to desiccation and deathdiffer between nitrogen and carbon dioxide, and nitrogen is generally moreeffective, the operating conditions and run times for carbon dioxide fumigationcan be adjusted to achieve total disinfestation. Most treatments are done in rela-tively large, reusable commercial bubbles developed by Rentokil. Carbon dioxidefumigation has not found application with small bags or pouches. The earliestuse of carbon dioxide fumigation was in England in the late 1980s, but its con-tinued use there is deterred by regulations requiring that treatment be doneonly by licensed fumigators. Institutions in the United States with large collec-tions of furniture and other decorative arts—such as the Henry Francis duPontWinterthur Museum in Winterthur, Delaware; Old Sturbridge Village inSturbridge, Massachusetts; and the Society for the Preservation of New EnglandAntiquities in Boston—have quickly moved to put carbon dioxide units on anoperating basis. This technology is also being applied to safeguard the naturalhistory collections at the Smithsonian Institution in Washington, D.C. In Canadathe use of carbon dioxide was promoted by Thomas Strang of the CanadianConservation Institute and was adopted by a number of Canadian museumssuch as the National Museum of Science and Technology in Ottawa and theCanadian Museum of Civilization in Hull. In Germany professional fumigatorGerhard Binker employs carbon dioxide on a massive scale to treat entire struc-tures, primarily churches, that are either tented or leak-sealed.

This book contains many references to the published literature on anoxia andfumigation. They are indicated in the text with parentheses showing the authorand year of publication. Complete publication information is listed in the Refer-ences. Unlike most scholarly texts, and because this book is intended to beeminently practical, there is also an unusual number of informal descriptions ofactual conservation experience that either have not been published or were pub-lished in limited-circulation periodicals. Most of these descriptions were directlycommunicated to the authors by letter or interview, in which case there is noparenthetical reference after the individual or institution is named.

xiv

Acknowledgments This has been a strange monograph to compile because the information camefrom three quite different sources: from the literature, from the experiencegained over a decade of research on anoxia at the Getty Conservation Institute,and from discussions and interviews with conservators and conservation scien-tists who have developed and used inert gas procedures. To those conservatorsand conservation scientists I owe an enormous debt of gratitude. They have pro-vided the data that are the basis for about two-thirds of this book. My apprecia-tion goes out to Nieves Valentín, John Burke, Mark Gilberg, Steven Price, VinodDaniel, Thomas Strang, Gordon Hanlon, Kerstin Elert, Sue Warren, Lucy Cipera,Martha Segal, Rebecca Snetselaar, Madeleine Fang, Steve Colten, Lesley Bone,David Casebolt, Dale Kronkright, Alan Johnston, Mark Roosa, Jeremy Jacobs,and Gary Rattigan.

Special acknowledgments go to John Burke, Steve Colton, Mark Gilberg, GordonHanlon, Alan Johnston, Dale Kronkright, Steven Pine, and Nieves Valentín, withwhom I had particularly lengthy and informative discussions. They responded tomany requests for additional information, drawings, and photographs and pro-vided valuable review and commentary on portions of the book. Those profes-sionals as well as Shin Maekawa, Vinod Daniel, Jeremy Jacobs, Gary Rattigan,Thomas Strang, and Sue Warren will continue to earn my gratitude for havingagreed to act as resources who may be contacted and queried on how to carryout anoxia and carbon dioxide fumigation. For sharing their expertise with me, Iwant to thank Gerhard Binker, Michel Maheu, John Newton, and Colin Smith, allprofessional fumigators who are experienced in the procedures described in thisbook. They also are willing to answer questions.

In addition, I want to thank Jonathan Banks of the Australian CommonwealthScientific and Industrial Research Organization for information on grain storagestudies using modified atmospheres for insect control, and Michael Rust of theUniversity of California, Riverside, for background and data on the biology ofinsects and for his review of chapters on this topic. The technical information onthe development of anoxia methodology and research on selecting the bestmaterials and methods of monitoring came primarily from my coauthor, ShinMaekawa, and his research fellow, Kerstin Elert. Her dedicated and resourcefulhandling of the battery of scientific questions connected with this project con-tributed greatly to its success.

The preparation of many drafts was done by Karen Sexton-Josephs at the startof this project and in the final stages by Elaine Holliman, both of whom workedwith dedication and skill. However, for the bulk of the effort in constructing thebook, I must thank Ford Monell, who prepared most of the chapters and verycreatively turned raw data and sketches into tables and figures. I want to thankmy colleague James Druzik for many helpful discussions and a review of thecompleted manuscript. Last, and yet foremost, I offer my deepest gratitude toFrank Lambert for a highly critical review of the text that suggested many sub-stantial and beneficial changes.

—Charles Selwitz

xv

DesiccationChapter 1

Mechanisms ofInsect Mortality

A number of mechanisms have been proposed to show how oxygen deprivation(anoxia) causes increased mortality, but desiccation seems to offer the bestexplanation for the results obtained with the procedures and conditions used toeradicate museum pests. The evidence is threefold. First, insect physiology pro-vides a well-defined respiratory system that leads to accelerated desiccation inthe absence of oxygen or in the presence of modest amounts of carbon dioxide.Second, death rates generally increase as conditions move in directions favoringdehydration. Thus, increasing temperature or decreasing humidity typicallymakes anoxia a more effective killing procedure. The third important line of evi-dence is that mortality rates are positively associated with weight loss whichunder anoxic conditions can occur only by loss of water.

Insects are able to control two important processes—the exchange of oxygenand carbon dioxide, and the conservation of water—by a series of orificesknown as spiracles (Fig. 1.1), which are part of a gas-transport system. A waxyepicuticular layer on the spiracles prevents water loss and maintains low oxygenpermeability, thus putting the spiracles in control of the supply of these essentialsubstances. Spiracles lie along the abdomen and thorax, where they are attachedto regulatory organs (Fig. 1.2). The spiracles control the access time for the trans-port of gases and the size of the pathway through the body wall to the trachea.Hassan (1943) suggested that the successful evolution of insects as a group ofterrestrial animals was largely facilitated by this regulatory system. The spiraclesare normally kept closed to minimize water loss and are opened just enough forthe insect to take in needed oxygen. When oxygen is scarce, however, they areforced to open more frequently and more widely, thus causing dehydration. Aninsect must get rid of carbon dioxide as well, and high concentrations of this gasin modified atmospheres, quickly sensed when the spiracles open, will also leadto sustained opening and, consequently, dehydration. These two conditions—very low oxygen levels and high concentrations of carbon dioxide—are therationale for the two different types of modified atmospheres developed for pestcontrol; both force the spiracles to open and remain open. This unnatural con-dition leads to high rates of water loss, as much as seven to ten times higherthan when the spiracles are closed (Mellanby 1934; Wigglesworth and Gillett1936; Bursell 1957; Jay, Arbogast, and Pearman 1971). Under conditions of high

1

Figure 1.1

Atriate spiracle with lip type of closing.Atrial oriface

Peritreme

Trachea

Tracheal oriface

Atrium

Figure 1.2

Location of spiracles on insect body.(By permission. From Merriam-Webster's

Collegiate® Dictionary, Tenth Edition, © 1997

by Merriam-Webster, Incorporated.)

carbon dioxide concentrations accompanied by high humidity, the role of dehy-dration in causing death is less clear. The firebrat is an insect that typically livesat high temperatures and high humidity. In this species, high carbon dioxideconcentrations relax the spiracle muscles, causing the spiracles to close by elasticrecoil. There is low tracheal water loss in the firebrat as exposure to carbon diox-ide brings on death (Noble-Nesbitt 1989). Donahaye (1990) also finds that othermechanisms must be invoked to describe the effects of high humidity and highcarbon dioxide atmospheres on red flour beetles.

Rising temperatures increase insect respiration, resulting in a greater productionand loss of water. This should increase mortality, and the work of Valentín(1993) demonstrates that it does. Figure 1.3a shows the minimum exposuretimes needed to achieve complete insect mortality at 40% relative humidity (RH)and 300 ppm oxygen in argon for eight different species at 20, 30, and 40 °C.Figure 1.3b provides the same comparison with nitrogen as the carrier gas. Foreach carrier an increase in temperature from 20 to 30 °C decreased the exposuretime approximately 30%; an increase from 20 to 40 °C shortened the time about90%. There was an even more dramatic decrease in exposure time for the mostresistant species, the old house borer. Jay, Arbogast, and Pearman (1971)showed the relationship between mortality and RH by examining the deathrate of red flour beetles and confused flour beetles in nitrogen atmospheres con-taining between 8000 ppm and 10,000 ppm oxygen after 24 hours at relativehumidities of 9%, 33%, 54%, and 68%. As shown in Figure 1.4, both speciesshowed a marked increase in mortality as the RH decreased. Decreasing the RHenhances the anoxic effect by increasing water transport and loss through openspiracles. When total body-water loss approaches 30%, most insects die.

Under anoxic conditions, any decrease in weight is due almost entirely to loss ofwater. In situations where there is a positive association between kill rate andweight loss, the mechanism responsible is highly likely to involve dehydration.Thus weight-loss studies are important for explaining a mechanism of mortalitybased on desiccation. One of the few such studies was done by Navarro (1978).For red flour beetle adults and Ephestia cautella pupae, he found that as thelevel of oxygen was uniformly lowered in 1% increments below 4%, the time toreach 95% mortality also fell in a constant manner. In the same study, however,the death rate for the rice weevil did not increase in a uniform way as the oxy-gen concentration was reduced in 1% increments from 3% to 0%. Navarrofound instead a small dip in the time required for 95% mortality at 1% oxygenand surmised that this may be due to differences in spiracular movements at thisparticular concentration. He further determined the daily percentage weight lossfor the rice weevil at each oxygen level and found that it was highest at 1%. Spi-racles open as oxygen concentration falls, but this apparently reaches a maxi-

2

Figure 1.3a,b

Effect of temperature on the minimum expo-sure time to (a) argon and (b) nitrogenrequired to achieve complete insect mortalityat 40% RH and 300 ppm oxygen (Valentín1993). (Courtesy of Nieves Valentín.)

Exposu

re ti

me

(day

s)Ex

posu

re ti

me

(day

s)

Insect species(a)

Insect species(b)

mum at 1%. As the oxygen content continues to drop, the spiracles tend toclose. Thus, at 1% oxygen, as at other concentrations, desiccation is the causeof death, and there is no need to invoke an alternate mechanism.

There is a critical level of weight loss associated with mortality. Below this level,only weaker specimens succumb; above it, the general population rapidly per-ishes. Navarro found that for Ephestia cautella pupae, the critical range is from17% to 25% weight loss. Jay and Cuff (1981) measured weight loss and evalu-ated mortality for three life stages of the red flour beetle at 1% and 3% oxygen.To achieve mortalities of 90% or higher, weight loss had to reach about 250 mgfor twenty-five larvae and 150 mg for twenty-five pupae. Loss of weight wasslower at 3% than at 1% oxygen, but when enough time was allowed to bringthe weight loss to the critical level, the same high kill rate was achieved. This isadditional evidence that anoxia kills by causing dehydration.

There is decreased susceptibility of flour beetles and other insects to anoxia ashumidity increases. Studies showing a correlation between weight loss and mor-tality (Navarro 1978; Jay and Cuff 1981) at relative humidities of at least 54% inan anoxic nitrogen atmosphere suggest that the desiccation mechanism is still inplay. Dehydration of the insect occurs even when it is surrounded by high levelsof water vapor. The mechanism is twofold. First, nitrogen passes into the insectthrough its body wall, picks up water from the internal organs as it passesthrough the thorax and the abdomen, and then carries the water out of the

3Mechanisms of Insect Mortality

Chapter 1 4

Figure 1.4

Mortality of red and confused flour beetlesexposed 24 hours at 26 °C to 1 % oxygen innitrogen at four relative humidities (Jay,Arbogast, and Pearman 1971).

Perc

ent

mor

talit

y

Red flour beetle Confused flour beetle

insect through the spiracles. Second, water is kept out of the insect by the epi-cuticle, that is, the waxy, waterproof covering of the spiracles (Edney 1967).

Argon and helium are even more effective atmospheres for anoxia than is nitro-gen, and the explanation involves differences in the permeability of the insect'sbody wall to these gases. The previously cited work of Valentín, in addition todemonstrating a positive correlation between temperature and mortality rate,also compared the kill rates for eight insect species exposed to argon and nitro-gen atmospheres. Under comparable conditions (40% RH, 300 ppm oxygen at20, 30, and 40 °C), it generally took 50% longer with nitrogen than with argonto reach 100% mortality (Fig. 1.3). AliNiazee (1972) reported that helium alsoprovides a much faster kill than nitrogen. This investigator found that heliumgenerally took only half as long as nitrogen to achieve 97% mortality with redflour beetles and confused flour beetles (Table 1.1).

If these differences in effectiveness are due to the permeability of the inert gasesthrough pores in the body walls of the insects, then these values are in line withthe van der Waals radii of helium (1.22 Å), argon (1.91 Å), and molecular nitro-gen (2.31 Å). Thus the smallest atoms (helium and argon) move the fastestthrough the minute pores and dehydrate with corresponding speed; the largernitrogen molecule takes the longest.

Anoxia at High Humidities

A mechanism other than desiccation takes over at extremely high humiditieswhere anoxia is still effective, but slower. Donahaye (1990) found that loss ofwater from red flour beetles exposed to 0.5% oxygen in nitrogen at 95% RHoccurred only after the insects were dead. Obviously, a biological pathway otherthan dehydration has to be involved in this lethal process. Donahaye's research,however, was primarily concerned with a different question: Could species suchas the red flour beetle develop a resistance to modified atmospheres? He sub-jected the beetles to two modified atmospheres—0.5% oxygen in nitrogen, and20% oxygen and 15% nitrogen in carbon dioxide—at 95% RH until 30-50%remained alive. (The extremely high humidity was employed to suppress thedesiccation mechanism.) He repeated the treatment with the offspring of thesurvivors for over forty generations. Two resistant strains of beetle developed,each resistant only to the specific atmosphere to which it had been subjected.Compared to the original beetles, the resistant strains required nine times the

9% RH33% RH54% RH68% RH

Mechanisms of insect Mortality 5

Table 1.1

Effect of temperature and inert-gas type on theexposure time required to produce 97% insectmortality at 38% RH and 0.05% oxygenconcentration (AliNiazee 1972).

Temperature

15.6 °C 21.1 °C 26.7 °C

Confused flour beetleHeliumNitrogen

Red flour beetleHeliumNitrogen

12 h30 h

12 h

24 h

9h

15 h

9h

13.5 h

5h

12 h

6h

12 h

exposure before a 50% mortality level was observed. Additionally, insectsremoved from the modified atmospheres after thirteen generations retained83% of their resistance after eight generations of exposure to ambient condi-tions. The type of compensations that enabled the red flour beetles to surviveincluded a decrease in respiration rate, an increase in stored oxygen reserves,physiological changes to prevent water loss, and other biochemical adaptations.

The finding that strains of insects resistant to anoxia readily arise in high humidi-ties raises concerns about generating resistant species both in natural environ-ments and in the course of practicing pest control in museums and libraries.Insects, after all, are believed to have descended from aquatic ancestors thatlived under anoxic conditions. A number of larvae and adult insects that are nowat home in aquatic surroundings are recent returnees, like ocean-bound mam-mals, and they have at times readapted to anaerobic respiration. Insect survivalwithout oxygen over extended periods of time is rare but not impossible, andthere are examples of transitions from an aerobic to an anaerobic mechanismwhen the available oxygen drops below a critical level. When this happens, theenergy-generating mechanism that sustains life causes the disappearance ofbody carbohydrates (in the form of glycogen) and fat, and the accumulationof ethanol by fermentation accompanied by less than the expected amount oflactate (AliNiazee 1972). In polluted, nutrient-rich bodies of water, and particu-larly in ice-covered ponds, the oxygen level may fall to zero for periods ofweeks. Under these anoxic conditions, insect larvae living along the bottomdepend on anaerobic energy production; several types of water midges, forexample, are known to survive for longer than one hundred days in the absenceof oxygen.

That these and other hypoxic survivals occur when insects are immersed in wateror surrounded by ice suggests that, as in the Donahaye studies, evolution towardresistant species occurred when desiccation was overwhelmed by a supply ofwater. The evidence indicates that anaerobic ethanol formation in all of thesespecies came about as rather recent adaptations to their unique environments(Zebe 1991).

More common and less drastic adaptations have produced the museum peststhat take longest to eliminate by conventional anoxia treatment; these speciesare characteristically in environments tending toward confinement, dampness,and high carbon dioxide and low oxygen concentrations. Examples would begrain-storage insects, such as the cigarette beetle and dermestids, and insectsthat burrow deeply into books and wood. Still, they are amenable to anoxia; itjust takes a little longer. Always, 100% mortality must be the goal for effectivepest management. Total kill is far more important than trying to save time byshortening the treatment.

6Mechanisms of insect Mortality

Chapter 2

Anoxia as aConservationProcedure

Separate publications in the same year by Bailey and Gurjar (1918) and Dendy(1918), on the use of airtight storage to rid grain supplies of insect pests,marked the beginning of scientific inquiry into anoxia. The 1950s saw increasedinterest in and development of modified atmospheres for food preservationbecause of greater awareness of the environmental dangers posed by pesticides.An important stimulus was action by the U.S. Environmental Protection Agencyin 1980 approving the use of nitrogen and carbon dioxide for all agriculturalproducts. The degree of attention and development given this technology wassummarized in a major review by Calderon and Barkai-Golan (1990). This meth-odology, however, is not widely used for preserving foodstuffs in large-scalecommercial operations because of the expense of long treatment times and thedevelopment of safer procedures using methyl bromide and phosphine.

From Food Preservation to Conservation

Custodians of cultural property also have had a long history of concern aboutthe damage done by insects. In the 1980s, as interest in modified atmospheresfor food preservation peaked, conservation scientists began to study how thistechnology could be adapted to museum needs. Fumigation with toxic gases isobviously not a safe option within public buildings, but the long treatment timesrequired for nontoxic gases are generally acceptable. Although a carbon dioxideatmosphere had been favored for the preservation of foodstuffs, conservatorssaw more advantages in using nitrogen with low oxygen concentrations to treatcultural property because anoxia provides a higher degree of inertness and iseasier to establish for small-scale operations. But the transfer of procedures fromfood preservation to museum use did not come easily; some important lessonshad to be learned first.

Early recognition that modified atmospheres could be used to control pests forconservation purposes appeared in Approaches to Pest Management in Muse-ums, a 1985 book by Keith Story. He wrote, however, that for practical considera-tions, modified atmospheres should be based on carbon dioxide rather than onnitrogen containing essentially no oxygen. Story did not describe any use ofmodified atmospheres other than those involving very large volume contain-ments. Indeed, large-scale work is more readily done with moderate (50-60%)levels of carbon dioxide rather than with nitrogen having very low concentra-tions of oxygen because there is always the possibility of air leaks in large cham-bers. According to Story, such levels of carbon dioxide at 60% RH and 21 °C willkill all life stages of most insects in 4 days. Earlier, Arai and Mori (1980) had ex-amined anoxia as a means of simultaneously treating fungal growth and insectinfestation in cultural property. It took 20 days to achieve a 92% kill of rice wee-vils with nitrogen containing 1000 ppm to 2000 ppm (0.1-0.2%) oxygen innitrogen at 55-70% RH. The fungal growth, on the other hand, was inhibitedbut not killed. This led the authors to recommend fumigation with insecticidesfor a more rapid extermination. However, their description of variations in theoxygen concentration in the laminated polymer bag they used suggests that thebag was excessively permeable to oxygen. In 1990 deCesare described in alibrary newsletter how books can be rid of insect pests by holding the worksunder an argon atmosphere in plastic bags or trash cans. Unfortunately, he gaveno details on critical aspects of treatment such as insect types, oxygen concen-tration, temperature, length of treatment, or level of mortality achieved. Thework by deCesare pointed in the right direction, but its qualitative nature did notoffer a procedure that could be used by conservators with any assurance.

Conservation scientists at two separate institutions, the Australian Museum inSydney and the Getty Conservation Institute in Los Angeles, first recognized andworked through these problems. Mark Gilberg joined the Conservation Divisionof the Australian Museum in the spring of 1987. His first assignment was to find

7

Chapter 2 8

a substitute for ethylene oxide as a fumigant to rid museum objects of insectpests. At the time, there was considerable interest and activity in Australia inusing modified atmospheres for suppressing insect infestations in grain storage(Annis 1987), and this appeared to be a promising direction for the scientists atthe Australian Museum. Gilberg was able to enlist the help of Jonathan Banks,senior principal research scientist with the Stored Grain Research Laboratory inthe Commonwealth Scientific and Industrial Research Organization. He hadbeen involved in research on modified atmospheres for insect control for manyyears, with both low-oxygen systems and high carbon dioxide atmospheres(Banks 1979; Banks and Annis 1990). Banks quickly saw how the technologydeveloped for grain storage could be adapted to the preservation of culturalproperty and made a number of valuable suggestions. He recommended thenitrogen anoxia approach because he felt that insects were more likely to buildup a resistance to the toxicity of a carbon dioxide atmosphere. Banks alsoacquainted Gilberg with the properties of oxygen-impermeable plastic films forcontainment and with the use of commercial oxygen scavengers. The body ofdata that had been compiled over fifty years of food-preservation studies bymany researchers was an invaluable information resource for museum personnelwho were adapting anoxia to their needs. However, Banks's intercession andinstructions provided an immediate, direct transfer of this technology betweenthe two disciplines, saving conservation scientists several years of research effort.

Early Museum Anoxia Studies

Two years after joining the Australian Museum, Gilberg published the first reportdescribing effective control of insects by anoxia in a manner relevant to museumneeds (1989a). He studied cigarette and drugstore beetles of the Anobiidaefamily and carpet beetles of the Dermestidae family, which are common museumand grain storage pests, as well as powderpost beetles and webbing clothesmoths. Gilberg maintained insect colonies in glass containers at 0.4% oxygenconcentration, 65-70% RH, and a slightly elevated temperature of 30 °C. Heachieved 100% mortality with all species and life stages in 7 days. Gilberg con-sidered this temperature, selected to shorten the required exposure time, to bethe maximum safe level for museum objects. He also noted that studies with therice weevil showed that this anoxia-resistant grain storage pest was completelykilled in 1% oxygen at 29 °C in 3 weeks, and he suggested that this length oftime might define an upper exposure limit for museum work. Gilberg (1991)repeated his work with these insects to confirm that 3 weeks of exposure wouldreliably provide total kill of all life stages. He further promoted the use of chemi-cally reactive oxygen absorbers in enclosed systems as a convenient and effec-tive way of bringing oxygen concentrations down below 1% (1990). Gilberg'swork has led to the widespread use of an oxygen scavenger based on a highlyreactive, high-surface-area ferrous compound that is commercially packagedand sold by the Mitsubishi Gas Chemical Company under the Ageless tradename. Details on using a scavenger to maintain a very low oxygen level arediscussed in chapter 3.

When Gilberg started his anoxia studies, work was proceeding at the Getty Con-servation Institute on the design and construction of nitrogen-filled, hermeticallysealed display and storage cases, initially for holding the Royal Mummies of theEgyptian Antiquities Organization that are now on display at the EgyptianMuseum in Cairo. It was hoped that by drastically reducing the level of oxygenin the cases, and then by controlling it as well as the RH, museums would have acost-effective way to prevent deterioration of extremely old organic objects suchas mummies. Such objects are affected by changes in humidity, biological attack,thermally and photochemically induced oxidation, and gaseous and particulatepollutants. The prototype cases developed by the Getty Conservation Institutewere built so that under normal conditions they could be maintained at less than

Anoxia as a Conservation Procedure 9

2% oxygen concentration without major maintenance for up to ten years(Maekawa, Preusser, and Lambert 1993).

When Nieves Valentín joined the Scientific Program of the GCI in 1987, she wasasked to evaluate the effectiveness of nitrogen anoxia for deterring the biologi-cal deterioration of ancient artifacts. Work presented in August 1989 (Valentínand Preusser) described studies in which colonies of bacteria and fungi on parch-ment were exposed to an atmosphere of 1% oxygen in nitrogen at 33-43% RHfor 3 weeks. Both fungi and bacteria were deterred from further growth. Theeffect on insects was more dramatic: nitrogen anoxia eliminated all stages of thelife cycle of the fruit fly. The exposure time required to do this decreased asthe RH fell, but it was inversely related to temperature. At room temperature,75% RH, and 5000 ppm oxygen, all life stages, including eggs, were killed in80 hours. This time requirement was only 30 hours at 30 °C and 40% RH. Addi-tionally, studies with a common museum pest, the drywood termite, demon-strated that infestations in wood could be totally eliminated by containment inaluminized plastic bags filled with nitrogen containing less than 1% oxygen at25 °C and 40% RH.

Determining Kill Times

These initial studies showed that by diligently maintaining a nitrogen atmo-sphere with a very low oxygen content, a complete kill of all life stages for somepests could be obtained in reasonable periods of time. A good start had beenmade in defining process requirements. Still, there were a number of materialproblems to be solved before general procedures could be developed that wereeasy and effective for conservators to carry out in most museums. Also of para-mount concern was determining the minimum amount of time required for anassured 100% mortality rate for the most destructive museum pests and therelationship of this time to operating parameters. This assured kill time is criticalfor defining both the cost and usefulness of anoxia treatment. For example, thecost of monitoring oxygen concentration and maintaining airtight chambers andtents, two very important practical matters, can be reduced significantly if mor-tality rates at higher oxygen concentrations are known. These considerationshad been examined for grain storage problems much earlier. In 1977 Banks andAnnis, noting that the rice weevil was the most tolerant grain pest in low-oxygen atmospheres, reported the exposure times required for the complete killof this species in 1% oxygen in nitrogen over a range of temperatures from15 to 35 °C. An exposure of 6 weeks was recommended at 20 °C, but this fellto 2 weeks at 30 °C. Other pest groups, such as Tribolium, Oryzaephilus, andRhyzopertha, required shorter exposures, perhaps half as long. However, ifthere were any doubt about the species present, it was suggested that thelonger times be used. The addition of carbon dioxide appeared to allow someshort, but undefined, reduction in exposure time.

The difficulty of answering questions about treatment times for art objects wasfirst raised by Gilberg in 1990 and subsequently addressed in a study by Rustand Kennedy in 1993 (also Rust et al. 1996) that was sponsored by the GettyConservation Institute. In selecting the pest species for their study, Rust andKennedy used surveys from natural history and art museums that indicated thatbeetles belonging to the families Anobiidae and Dermestidae and moths belong-ing to the family Tineidae are major pests. Tables 2.1 and 2.2, taken from theirreport, summarize the common insect pests found in museums in North America.The twelve species that they studied were chosen primarily from these tables.Several other important species were not included in the study because theycould not be legally brought into the United States. For this work, an anoxicatmosphere at 55% RH was produced by bubbling a sidestream of dry, prepuri-fied nitrogen containing 20 ppm oxygen through water and then recombining

Table 2.1

Insect pests found in North Americanmuseums.a

Order Family Species Common name

Coleoptera

Lepidoptera

Dictyoptera

Thysanura

Anobiidae

Bostrichidae

Cleridae

Dermestidae

LyctidaeTenebrionidae

Tineidae

Blattellidae

Lepismatidae

Lasioderma serricorneStegobium paniceum

Anobium punctatumXestobium spp.

Dinoderus minutus

Necrobia rufipes

Anthrenus flavipes

Anthrenus verbasci

Attagenus megatoma

Dermestes lardarius

Thylodrias contractus

Trogoderma indusum

Lyctus spp.

Tribolium confusum

Tinea pellionella

Tineola bisselliella

Blattella germanica

Supella Iongipalpa

Lepisma saccharina

Thermobia domestica

Cigarette beetleDrugstore beetle

Furniture beetleDeathwatch beetle

Bamboo powderpost beetle

Redlegged ham beetleFurniture carpet beetle

Varied carpet beetleBlack carpet beetle

Larder beetleOdd beetle

Cabinet beetle

Powderpost beetleConfused flour beetle

Casemaking clothes moth

Webbing clothes moth

German cockroach

Brownbanded cockroach

SilverfishFirebrat

aCompiled from Kingsolver(1981) and Beauchamp, Kingsolver, and Parker (1981) as referenced by Rust and Kennedy(1993).

wet and dry streams with appropriate metering as described in chapter 4. Twelveinsect exposure chambers built of quarter-inch-thick acrylic sheets (Fig. 2.1),each 1.5 ft3 in capacity, were flushed with this gas. Each was an independentexperimental chamber containing two packets of the commercial scavenger Age-less Z-1000, to maintain a very low concentration of oxygen, as well as a smallpan of aqueous saturated magnesium nitrate solution, to ensure an RH of 55%.A Teledyne trace-oxygen analyzer with a fuel-type sensor was used to determinethe oxygen level. Colonies of all life stages of the selected species were thusmaintained at 25 °C and 55% RH in a nitrogen atmosphere containing less than1000 ppm oxygen.

The kill rates were first determined for an expanded series of time periods likelyto be long enough to provide total mortality. They were then reexamined using a

Table 2.2

Survey of natural history museums of NorthAmerica.a

Percentage of total respondents

Pests Pests encountered Greatest threat

Dermestid beetlesSilverfish

CockroachesAnobiid beetlesTenebrionid beetles

Mites

MothsUnspecified

7030261312

4

0

30

49

2

3

62

11

Chapter 2 10

aCompiled from Bell and Stanley (1981) as referenced by Rust and Kennedy (1993).

—

—


smaller number of exposure periods based on the time in the first series where100% mortality of all life stages occurred. For example, the initial screening ofthe webbing clothes moth involved five exposure periods of 3, 24, 48, 72, and96 hours. The 48-hour period appeared to provide 100% kill of all eggs, and inthe subsequent confirming study, exposures were limited to 48, 72, and 96hours. However, in one set the 72-hour period produced only 98.7% mortalityof the cocoon stage. An exposure of 96 hours produced a total kill of all stages,and this was selected as the recommended time for the webbing clothes moth.In the case of the cigarette beetle, the only time selected for the second-stageevaluation was 192 hours, which equaled the minimum time required to obtain100% mortality in the first screening, plus 6 hours. The 192-hour period wasshown to provide complete kill in four separate tests.

Results obtained by Valentín, Gilberg, and Rust and Kennedy are summarized inTable 2.3. This shows that overall the time required to kill 100% of the insectsvaried among the species and especially among the developmental stages of agiven species. Many of the pest species were completely exterminated with lessthan 72 hours of exposure, while the eggs of the cigarette beetle required 8 daysto guarantee 100% kill. The addition of carbon dioxide to the anoxic gas slightlydecreased the exposure needed for total kill, but increasing the temperature wasmore effective in reducing kill time than was adding carbon dioxide. In morerecent work, Rust and Kennedy (1995) examined the effect of oxygen concen-trations greater than 1000 ppm on the length of treatment for cigarette beetlesand furniture carpet beetles. They found that within this limited sampling,results were moderately dependent on the species. Although the minimumassured kill time for the cigarette beetle is relatively long—8 days at 1000 ppmoxygen in nitrogen at 55% RH and 25 °C—this increases relatively slowly as theoxygen concentration is increased to 6200 ppm. In contrast, furniture carpetbeetle larvae, which is the resistant stage for this species, require only 3 days forcomplete kill at 1000 ppm oxygen under these conditions, but this rises rapidlyto 20 days at 6200 ppm oxygen (Fig. 2.2). Above this concentration of oxygenin nitrogen, minimum assured kill times rise sharply for both species. These datasuggest that at the defined conditions of temperature (25 °C) and humidity(55%), timely and effective killing of pests by anoxia can be achieved at3000 ppm oxygen in nitrogen, but not much higher.

Figure 2.1

Insect exposure chambers.

Table 2.3

Minimum time recorded for assured 100% kill of all life stages of common museum pests.

Species

Oxygen

(ppm) Inert gas

Temp.

(°C)

%Relative

humidity

Time

(h) Reference

American cockroach

Black carpet beetle

Book beetle

Brownbanded cockroach

Cabinet beetle

Cigarette beetle

Confused flour beetle

Drugstore beetle

Drywood termite

Firebrat

Fruit fly

Furniture beetle

Furniture carpet beetle

German cockroach

Larder beetle

Longhorn borer beetle

Powderpost beetle

Rice weevil

Varied carpet beetle

Webbing clothes moth

Western drywood termite

<1000

300

300

300

300

<1000

<1000

4200

<1000

300

300

<1000

300

300

4200

10,000

<1000

5000

300

300

<1000

<1000

<1000

300

300

300

300

<1000

4200

10,000

10,000

4200

<1000

4200

<1000

Nitrogen

Nitrogen

Argon

Nitrogen

Argon

Nitrogen

Nitrogen

Nitrogen

Nitrogen

Nitrogen

Argon

Nitrogen

Nitrogen

Argon

Nitrogen

Nitrogen

Nitrogen

Nitrogen

Nitrogen

Argon

Nitrogen

Nitrogen

Nitrogen

Nitrogen

Argon

Nitrogen

Argon

Nitrogen

Nitrogen

Nitrogen

Nitrogen

Nitrogen

Nitrogen

Nitrogen

Nitrogen

25

30

30

30

30

25

25

30

25

30

30

25

30

30

30

22

25

30

30

30

25

25

25

30

30

30

30

25

30

20

26

30

25

30

25

55

40

40

40

40

40

55

65-70

55

40

40

55

40

40

65-70

40

40

75

40

40

55

55

55

40

40

40

40

55

65-70

12

12

65-70

55

65-70

55

120

72

48

72

48

72

120

168

192

144

96

96

144

96

168

360

48

80

168

120

72

24

96

240

168

120

72

120

168

1000

500

168

96

168

96

Rust and Kennedy (1993)

Valentín(1993)

Valentín (1993)

Valentín (1993)

Valentín (1993)



Gilberg(1989a)


Valentín (1993)

Valentín (1993)


Valentín (1993)

Valentín (1993)

Gilberg(1989a)

Valentín and Preusser (1989)


Valentín and Preusser (1989)

Valentín (1993)

Valentín (1993)




Valentín (1993)

Valentín (1993)

Valentín (1993)

Valentín (1993)


Gilberg(1989a)

Banks and Annis (1977)

Banks and Annis (1977)

Gilberg(1989a)


Gilberg(1991)


Chapter 2 12

The second largest group in Table 2.3 was provided by Valentin (1993), who sur-veyed museums and archives in northern Europe and the Mediterranean area todetermine the insects most prevalent in historic collections. The old house borer(Hylotrupes bajulus, which is known by several other common names, includinglonghorn borer beetle) and the furniture, drugstore, deathwatch, and black car-pet beetles were found in cellulose materials in museums, while the furniture,drugstore, powderpost, and book beetles were isolated from books and bundlesof documents taken from archives. The cigarette beetle was found frequently inplant collections. Samples of different life stages of each species were placedwith wood or paper in low-permeability, poly(vinylidene chloride) bags andpurged with pure humidified nitrogen or argon for 8 hours. Separate sampleswere held at 20, 30, and 40 °C for periods of 1-30 days. In all cases the oxygen


Figure 2.2

Effect of oxygen concentration on minimum

time for assured 100% mortality of cigarettebeetle and furniture carpet beetle at 25.6 °Cand 55% RH (Rust and Kennedy 1995).

Day

s re

quire

d fo

r10

0% a

ssur

ed m

orta

lity

Volume % oxygen in nitrogen

level was held between 200 ppm and 500 ppm. Following treatment, all sampleswere exposed to air under ambient conditions for seven months to disclose anysign of life. Valentín's results were in agreement with those of Rust and Kennedy.All life stages of the furniture and powderpost beetles, including eggs, wereeliminated in 3-5 days with 300 ppm oxygen in nitrogen at 30 °C and 50% RH.Rust and Kennedy found that exposure times of 72 hours or less were sufficientto treat infestations of the furniture carpet beetle and the powderpost beetle.Both groups found that the cigarette beetle was more tolerant of anoxia andrequired longer exposure. Rust determined that this species was eradicated onlyafter 8 days at 25 °C and 50% RH. As would be expected, Valentín found thatlonger exposures were needed at a lower temperature, for example, 9 days at20 °C and 40% RH.

By using argon instead of nitrogen, the required exposure time is decreasedabout 30%, but a modest increase in temperature—going from 30 to 40 °C—will lower treatment time as much as 85-90%. Valentín determined that thecomplete kill of all life stages of the species she studied could be achieved in lessthan 24 hours by operating at 40 °C. Although much shorter treatment timesresulted at elevated temperatures, little advantage is taken of this in currentpractice. Most treatments are done at room temperature, typically between20 and 25 °C. Gilberg and Roach (1992) describe a program of treatment at theAustralian Museum where objects were kept in bags under nitrogen in a thermalcabinet at 30 °C. Except for the Valentín study, there is little information aboutwork done at temperatures as high as 40 °C.

Thermal Techniques

There is a question of how high a temperature conservators can safely use in dis-infestation treatments. Operating above ambient temperature requires moreequipment and attention, but the principal reason for not doing treatmentsabove 30 °C is concern for the well-being of objects. However, much of this fearmay be without scientific justification. There is a growing body of literature sug-gesting that a large portion of general museum collections can be heated totemperatures approaching 60 °C without fear of damage, if adequate moisture ismaintained (Strang 1992, 1995; Child 1994; Thomson 1995; Ertelt 1993). Thishas created an advocacy for the disinfestation of cultural property by strictlythermal means. Strang has shown that for a wide range of cellulosic materials,the risks from thermal techniques are eliminated by ensuring that the heatingprocess does not drive water out of the object. Adequate humidification is

Larvae of the furniturecarpet beetle

Eggs of the cigarettebeetle

essential and is accomplished by enclosing objects in vapor barriers within thecontainment system. The Thermo Lignum Company sells equipment to do thisand has defined the procedures it believes are required for the safe thermal dis-infestation of museum objects. This is done by placing objects in a 4 × 3× 2.7 mchamber and slowly bringing the temperature to 52 °C. Even though most pestspecies are killed in one hour at this temperature, the treatment cycle takesabout fifteen hours because the unit is deliberately heated and cooled slowly. Inparallel, moisture is first added to and then removed from the chamber to keepthe RH essentially constant. Undesirable thermal side effects in wood, such asshrinkage, cracking, and warping, are eliminated. The Thermo Lignum unit isalso designed to operate in an atmosphere of 10% carbon dioxide or in nitro-gen containing only 1000 ppm oxygen. If thermal treatment is safe for objectsnormally maintained in air, it should not pose a threat to them under anoxicconditions.

There are materials whose exposure to temperatures up to 60 °C should be con-sidered with great caution; they include objects coated with soft waxes, ethno-graphic pieces containing deteriorated collagen, and animal specimens in naturalhistory collections with body oils and fats that will migrate when warmed. Theresults of work with thermal techniques should encourage conservators, who arethoroughly aware of which objects are prone to heat damage, to conduct anoxiatreatments at 30 °C and even somewhat higher, but they should be alert to theneed for careful moisture control.

Killing Burrowed Insects

There is natural concern that insects that have burrowed deeply into woodenobjects will be slow to respond to an anoxic atmosphere, and that this couldsubstantially increase the amount of time needed to reach 100% mortality. Oneapproach to determining the additional treatment time is to measure the diffu-sion exchange rates of nitrogen, oxygen, and air through different types ofwood. The porosity (air volume) of most woods when dry is about 45-50%.After a wooden block that has been standing in air is placed in a sealed chambercontaining essentially pure nitrogen, the oxygen content within the chamber willrise as the oxygen diffuses out of the pores. The amount of time needed to bringthe wood to anoxic conditions throughout can thereby be determined and thenused to calculate the additional time required for treating infestation. Lambertand Maekawa evaluate the exchange rate through poplar, oak, and walnut inthis manner (Lambert and Maekawa, forthcoming). Wooden blocks (3.8 X3.8 x 50.4 cm, about 700 cm3), both uncoated and coated along the sides withshellac, were placed in a 32 l acrylic chamber containing 5000 ppm oxygen innitrogen. The increase in oxygen was followed with a Teledyne trace-oxygenanalyzer, which could detect changes to 1 ppm. The time required to get toequilibrium for the six samples, described in Table 2.4, ranged from 9 hours foruncoated poplar to 120 hours for coated walnut. However, the long exchangetimes are misleading because they were determined for blocks of fresh wood.Insect boring and tunneling open up the structure of wood and dramaticallyincrease the rate of exchange. The time needed to get the oxygen out of

Table 2.4

Time required to completely displace air withnitrogen in dried wooda (Lambert andMaekawa, forthcoming).

Chapter 2 14

Wood type

Sample condition Poplar Oak Walnut

UncoatedCoated

9h

34 h

11 h

48 h

34 h

120 h

aWood block size: 3.8 x 3.8 X 50.4 cm; chamber size: 32 l; ambient temperature: 23 °C; starting 02% in N2:0.1-0.2%;equilibrium 02% in N2: 0.40%.


infested walnut, for example, fell to as little as one hour for untreated wood andfour hours for a heavily coated sample.

In a more direct assault on this problem, Rust and Kennedy (1993) sealedpowderpost beetles and western drywood termites inside blocks of wood andcompared the time needed for their total demise with the time needed to killunsealed insects. Beetles of the Lyctidae family, collectively referred to as truepowderpost beetles, feed on hardwoods such as oak, walnut, and mahogany.Western drywood termites, one of the most destructive termites in the UnitedStates, establish themselves in almost any wood that is not decayed. Rust andKennedy encased these insects in blocks of Douglas fir, measuring 8 x 8 x14 cm, that had been cut in half and hollowed in the center to hold small, open-ended, glass cylinders containing the insects. To eliminate air leakage throughthe sectioning, a sheet of parafilm was sandwiched tightly between the twoblock halves. There were 1.3-2.5 cm of solid wood surrounding each vial. Incontrol studies with exposed insects, it took 5 days to completely kill all lifestages of the powderpost beetle. With the beetles placed inside the wood, 100%mortality was achieved in 5 and 6 days in separate tests. Nearly identical resultswere obtained with the western drywood termites with and without enclosure inthe Douglas fir blocks. Rust and Kennedy concluded that the deep burrowing ofinsects into wood does not significantly increase the time needed for their com-plete extermination by anoxia.

Valentín (1993) investigated the exposure times needed to eliminate all insectinfestation in wooden pieces, books, and other items in a collection of museumobjects. Her results are summarized in Table 2.5. The species studied are activefeeders in works of art made of pine, cedar, walnut, oak, mahogany, and chest-nut. Column A shows the actual kill times for insects in the infested objects;column B shows the kill times for control insects exposed to comparable environ-mental conditions. There was little or no difference in required kill time betweeninsects in infested objects and their controls for six of the eleven treatments. Atthe other extreme, a complete kill of the furniture beetle in polychrome sculp-tures and a piano ranged from 10 to 14 days, substantially longer than the4 days needed for this species in the open.

Table 2.5

Environmental conditions used to achieve complete insect mortality in different historic objects exposed to an argon atmosphere (Valentín 1993).

Kill time (days)

Infested object size (cm) Insect species T (°C) % RH O2 (%)A

Insects in objectsB

Control insects

Books—35 x 25 x 16

Bundled documents—30 x 25 x 15

Sculpturea—200 x 80 x 52


Pianoa—200 x 100 x 100

Panela—80 x 30 x 15


Textiles—135 x 87 x 43

Panela—175 x 64 x 35

Framea—75 x 45 x 15

Plants

Drugstore beetle, powderpost beetle

Drugstore beetle, book beetle

Furniture beetle

Furniture beetle

Furniture beetle

Deathwatch beetle

Book beetle, furniture beetle

Black carpet beetle

Old house borer

Old house borer

Cigarette beetle

30

30

25

25

25

20

20

30

20

25

40

a Polychrome

40

65

45

45

40

50

50

45

50

45

35

0.02

0.02

0.04

0.03

0.03

0.02

0.02

0.03

0.04

0.03

0.03

4

4

10

10

14

7

10

7

15

10

1

3

3

4

4

4

5

4

2

14

10

1

16Anoxia as a Conservation Procedure

Chapter 3

Methods andMaterials

An anoxic environment suitable for treatment of infested objects can beachieved in a variety of ways. There are two basic approaches: static anddynamic. With the static procedure, the more common approach, objects areheld under high-purity nitrogen or argon in a tightly sealed container withas little transmission of gas as possible. The oxygen concentration is broughtdown to anoxic levels by one of three methods. The container is purged withmany exchanges of high-purity nitrogen; the oxygen is removed using largequantities of an oxygen absorber; or a combination of purging and absorption isused. With the alternate dynamic approach, an inert gas is continuously passedthrough the system during the treatment. Oxygen-free nitrogen or argon isused to flush all of the air out of the container by initially using a high purgerate; then, when an oxygen concentration of less than 1000 ppm is reached,the flow is reduced to that needed to maintain the low oxygen level for theduration of treatment.

The use of an oxygen absorber, by itself, to bring the system to anoxic condi-tions may be the simplest static procedure. The need for cylinders of nitrogenand humidification systems is avoided. The availability of premade treatmentbags with good oxygen-barrier properties and simple clamps that provide gas-tight seals should make disinfestation of cultural property by anoxia available atlow cost to almost any institution. This method is best suited for treating limitedquantities of small objects. More complex is treatment by nitrogen purging fol-lowed by sealing, with or without the additional use of an oxygen scavenger.This is the most widely used static method and applies to containments of allshapes and sizes, from small pouches to very large plastic tents as well as rigidmetal or plastic chambers, large and small. Humidification is generally used, butnot always. The dynamic method has found limited use in special situations, par-ticularly when the object is too large or too awkwardly located to be enclosed ina standard container. In this situation, containment is provided by shaping a bar-rier film around the object. Because unusual or uneven surfaces, such as floorsections, often become part of the containment, it is generally difficult to obtainthe required airtightness. However, gas leakage can usually be brought to a levellow enough so that a slow, continuous flow of oxygen-free gas through the sys-tem can be maintained effectively and economically. Humidification must beused with this dynamic approach. What is sometimes considered to be a thirdprocedure, called the static-dynamic method, is an intermediate approach wherethe purging gas is allowed to pass through the system very slowly.

A number of early workers in the field recognized the possibilities of these treat-ment methods and described and compared various approaches and systems inresearch papers. Getty workers (Hanlon et al. 1992; Daniel, Hanlon, andMaekawa 1993) described the characteristics of each method and first proposedthe terms static, static-dynamic, and dynamic. Koestler (1992) used a dynamicsystem to treat a sixteenth-century Andrea del Sarto panel painting that wasinfested with drywood termites. The panel, resting on a flat table, was coveredwith a large sheet of barrier film taped to the surface of the table. This contain-ment system was then flushed with humidified gas introduced through a port atone end and allowed to escape through another port at the far end. The treat-ment was successful but used forty cylinders of gas over twenty days. In thesame article, Koestler described three static-dynamic treatments. These used acontainment chamber fabricated from a utility storage cart made of heavy-gauge polyethylene; a Rentokil minifumigation bubble; and handmade pouchesof a barrier film made with an ethylene-vinyl acetate copolymer. Treatments withthe pouches combined the use of a nitrogen purge and oxygen scavengers.

Valentín (1993) also evaluated the efficacy of the static-dynamic method inthree different containment systems: plastic bags of low oxygen permeability; atiny (80 cm3), rigid vacuum chamber; and a 6.2 m3 fumigation bubble made of

17

Chapter 3 18

poly(vinyl chloride)-reinforced polyethylene. Humidification was used with eachsystem, but oxygen scavengers (thirty packets of Ageless Z-2000) were addedonly to treatments in the fumigation bubble. Valentín claims that all three sys-tems could provide satisfactory treatment.

Each variation in approach has its own advantages, disadvantages, and specialrequirements, but there are also critical needs in common. Each requires barrierfilms with low oxygen permeability; oxygen scavengers with high capacity andgood absorption potential; inexpensive nitrogen supplies with a very low oxygencontent; and effective methods of containment closure. These will be discussedin detail in the sections that follow.

Barrier Films

Most work with modified atmospheres for insect control is done today withflexible containers—that is, in pouches or bubbles made of plastic film. After ini-tial research using glass jars and rigid, thick-walled polymer boxes had demon-strated the feasibility of eliminating insect infestations using oxygen deprivation,polymer-film technology was ready, and products were available to make anoxiaa convenient procedure. Vapor-barrier laminates had been used for some timeon a vast commercial scale with perishables to seal in freshness and hold outoxygen and odorants and other causes of spoilage. From their earliest availabil-ity, these films were designed to be heat-sealable and good water barriers. JohnBurke (1992) has written a very useful review outlining the extent of the appli-cation of barrier films to conservation practice. He points out, for example, thatmuseums have been lining shipping crates with aluminized polyethylene (Mar-velseal 360) since the early 1980s. In 1986 the Oakland Museum sent on tour anexhibit of six hundred silver objects of local origin wrapped with this material.They were returned essentially untarnished after two years. Thus, vapor-barrierfilms had found their way into a wide range of conservation applications evenbefore they were considered for insect control.

To carry out anoxia treatments in plastic bags it is necessary to get the oxy-gen out and keep it out of the bags; this makes low oxygen permeability anextremely important film property. It is just as important that only one surface ofa film sheet soften sufficiently on heating to provide good adhesion to anothersurface and form a secure, gastight seal. Strength is also a critical property—thebag should be tear and puncture resistant and not likely to form pinholes. Poly-mer chemistry makes possible a high degree of property specificity for barrierfilms. This is achieved by the use of unique monomers; the correct design ofpolymerization conditions; and the appropriate physical processing of the films,for example, by annealing or stretch-orientation. However, there is a limit to theextent to which a combination of desirable properties can be built into a singlefilm. On the other hand, there is a synergism that is created when several filmschosen for specific values are sandwiched into a laminate. Barrier films used forpest control are all laminates of at least three single-composition polymer filmsthat impart specific properties, the most important of which are low oxygen per-meability, mechanical strength, and heat-sealability. Other properties, of varyingimportance, are low water-vapor transmission, transparency, printability, anti-static qualities, and fire resistance. Of course, low cost and availability are alsoimportant. Table 3.1 lists polymer films that are important for designing barrierlaminates for insect control and shows their oxygen permeability. Table 3.2shows how these films are combined into the barrier laminates that havebecome the most important for this work.

As seen in Table 3.1, polymers based on vinylidene dichloride and copolymers ofethylene and vinyl alcohol produce films of extremely low oxygen transmission.

Methods and Materials 19

Table 3.1

The composition and oxygen permeability of films.

Composition

Vinylidene dichloridemonopolymer and copolymerswith vinyl chloride

Ethylene-vinyl alcoholcopolymers (dry)

Ethylene-vinyl alcoholcopolymers (100% RH)

Nylon-6

Poly(ethylene terephthalate)

Poly(chlorotrifluoroethylene)

Poly(vinyl chloride)

Polypropylene (oriented)

Polypropylene (cast)

Polyethylene (low density)

Polystyrene

Polypropylene laminate withaluminum

Abbreviation

PVDC

EVOH

EVOH

NYL

PET

Aclar

PVC

PPO

PP

PE

PS

PPAL

Monomers

Vinylidene dichloride,vinyl chloride

Ethylene, vinyl alcohol

Vinyl alcohol, ethylene

Caprolactam

Ethylene glycol, terephthalic acid

Chlorotrifluoroethylene

Vinyl chloride

Propylene

Propylene

Ethylene

Styrene

Propylene, aluminum

Oxygena

permeability

0.16-2.46

0.11-0.80

8-16

40

56

141

200

7800

3700

4800

5200

3

Outstanding quality

Very low oxygen permeability



Low oxygen permeability, goodstrength

Low oxygen permeability

Low oxygen permeability

Low oxygen permeability, goodstrength

Heat-sealable

Heat-sealable

Heat-sealable

—Heat-sealable, low oxygenpermeability

aOxygen permeability in cm3 x mil/(m2 x days x atm).

More recent laminates have been developed using these polymers as core oxy-gen barriers. A second group of films in a somewhat higher—but still effective—oxygen transmission range consists of poly(vinyl chloride); poly(chlorotrifluoro-ethylene), which is known commercially as Aclar; nylon-6; and poly(ethyleneterephthalate). Films of nylon-6 and poly(ethylene terephthalate) are also toughwith high mechanical strength and are used primarily for these properties.Although pure Aclar has good, but not outstanding, oxygen-barrier values, com-posites containing this material do have very low oxygen permeability. However,high cost has been a deterrent to the use of laminates containing Aclar. Addi-tionally, some ambiguity has crept into the use of the term "Aclar." It properlyrefers to the polymer poly(chlorotrifluoroethylene), but a number of laminate

Table 3.2

Barrier laminates.

Core

(oxygen-barrier film) Compositiona Sourceb Trade name

Film thicknessc

(mil) Oxygen permeabilityd

PVDC

PVDC

Aclar

Aluminum

Aluminum

PET/PVDC/PE

PET/adh/PVDC/adh/PE

PET/PE/Aclar/PE

NYL/AI/PE

PPO/PE/AI/PE

1

2

3

3

4

Keepsafe (bags)

Filmpack1193

Film-O-Rap 7750Filmpack 1177Aclar

Marvelseal 360

Shield Pack Class A

2.5

4.9

4.5

5.3

5.0

7.1

0.28

50

0.01

0.01

aAdh = unspecified adhesive.b(1) Keepsafe Systems, Inc., Toronto, Canada; (2) Ludlow Corporation, Homer, La.; (3) Allied Signal Inc., Morristown, N.J.; (4) Shield Pack Inc., West Monroe, La.cMeasurements made at the Getty Conservation Institute.dOxygen transmission in cm3 x mil/(m2 x days x atm), measurements made at the Getty Conservation Institute.

Chapter 3 20

Table 3.3Comparative cost of barrier films.

Polymer

Laminates based on Aclar

Marvelseal 360

Filmpack1193

Cost(1996 US$/m2)

6.10-11.00

1.66-2.00

5.30-7.44

fabricators use this term broadly to designate composite barrier films containingpoly(chlorotrifluoroethylene). The polyolefins, polyethylene and polypropylene,are poor oxygen barriers but are incorporated into laminates for their neededadhesion and sealing properties. A laminate formulation with a flexible alumi-num film is included in Table 3.1 to show the effectiveness of this inexpensivematerial as a core barrier. Burke describes aluminum films as being formed ofvery small platelets that are deposited on top of each other to create anextremely long and difficult pathway for transiting gas molecules. Pinholesform easily in thin aluminum foil, so it is not a perfect solution.

Information on the chemical and physical properties of barrier films is important;it serves to educate and convince conservators of the large and significant differ-ences in oxygen permeability that occur as a function of film composition. This iswhy the selection of an appropriate polymer is critical. Common polyethylenebags, for example, are useless because of their high oxygen permeability. Con-servators who favor producing their own bags, or who must put together odd-shaped constructions out of barrier-film sheets and have to shop for the mostcost-effective laminates, need precise information on composition and proper-ties as well as on cost. The costs of three important commercial laminates arecompared in Table 3.3. For those who buy ready-to-use laminate bags, there isstill a need for this information in order to purchase wisely.

Gas Supplies

Nitrogen is widely available from a number of vendors in convenient, but heavy,metal cylinders. The Linde Division of the Union Carbide Company is a primeproducer of industrial- and research-grade gases, marketing them through a net-work of distributors. They offer a range of nitrogen grades: research, ultra-highpurity, prepurified, high purity, extra dry, oxygen free, and zero grades. Argon isprovided in about the same range of grades. The most suitable grades of nitro-gen for insect control are those designated as "prepurified" and "high purity,"which contain 5 ppm and 10 ppm oxygen, respectively (Table 3.4). A number ofcompanies, particularly in industrial areas, market cylinders of nitrogen suitablefor anoxia treatments under a variety of grade descriptions and will generallysupply written analyses providing the oxygen content for any grade. Industrial-grade nitrogen, usually offered as 99.7% nitrogen, is widely available and islower in cost than the research grades. Although its use may be limited to initialpurging stages, this grade often shows a surprisingly low oxygen content onanalysis (e.g., 20 ppm) and is certainly useful for treatments. For fumigationwith carbon dioxide, which is typically used at 60-70% for insect control, the99.7% concentration of the "bone dry" grade is quite adequate and provides arelatively low gas-volume cost. Other expenses associated with using gas cylin-ders, based on 1996 prices, are approximately US$240 for a regulator and a tankrental fee of about US$6 a month.

The use of larger containment systems requiring greater amounts of nitrogen willlead to the accumulation of a large number of cylinders during each treatment.Much of the awkwardness of handling a large quantity of tanks can be avoided

Table 3.4

Price comparison of argon, nitrogen, and carbon dixode.a

a Data and prices from Gilmore Cyrogenics, El Monte, Calif., February 1,1996.b At standard temperature and pressure.c Includes monthly rental for container.

Gas

Nitrogen

Nitrogen

Argon

Argon

Carbon dioxideLiquid nitrogen

Cylinder

size

T

T

T

T

K

—

Grade

PrepurifiedHigh purity

Prepurified

High purity

Bone dry

—

Purity

(%)

99.998

99.99

99.998

99.996

99.7

99.99-99.999

Weight(lb)

22

22

34

34

50

273

Oxygen

(ppm)

5

10

3

5

—

10

Water(ppm)

3

3

3

3

—

3

Gas volumeb

(ft3)

304

304

336

336

430

3495

Cost

(US$/cylinder)

46.00

37.00

65.00

50.00

15.00

90.00C

Approx.

cost

(US$/100 ft3)

15.00

12.00

20.00

15.00

3.50

2.60

Approx.

cost(US$/m3)

5.10

4.08

6.80

5.10

1.19

0.90

by connecting them in parallel on a manifold. Even so, working with five tothirty heavy cylinders can be expensive, physically demanding, and dangerous.This problem can be circumvented by using a nitrogen generator, which provideslarge volumes of oxygen-free gas at low cost after purchase of the generator.Units, such as the Prism generator manufactured by the Air Products and Chemi-cals Corporation, are available on a turnkey basis in many countries. The Prismnitrogen generator, model 1300, uses a membrane to separate high-purity nitro-gen from air and can supply 41 m3 of nitrogen with less than 100 ppm oxygenper day. It consists of an air compressor, a series of filters, and a trace-oxygenanalyzer in the output stream. The air is compressed and passed through anearly filter system to remove water, dust particles, and gaseous pollutants, andthen sent through the membrane filter where separation of the nitrogen streamoccurs. The system can produce nitrogen of 95-99.99% purity, although thehighest purity is achieved at the lowest flow rate. In 1996 the unit cost aboutUS$18,000 with a 230 VAC 3 phase motor/compressor and incurred expenses ofUS$200-US$300 per year for replacement filters and the sensor in the oxygenanalyzer. In studies at the Getty Conservation Institute, Maekawa and Elert(1996) found the Prism nitrogen generator easy to operate and reliable over tenweeks of continuous usage, although the compressor was somewhat noisy. Themodel was greatly oversized for the 10m3 bubble under test but it would bequite satisfactory for a 100 m3 chamber. Adding a pressurized storage tankwould enhance the cost-effectiveness of this system.

Another convenient and inexpensive source of large volumes of nitrogen gas isliquid nitrogen. The conventional 160 l tank of liquid nitrogen supplies 9.9 x 104 lof gas containing 20 ppm oxygen. This volume is equivalent to that provided bytwelve T-cylinders. The tank contains internal heat exchangers, which vaporizethe cold liquid and produce flow rates of more than 100 l min-1, although ratesof around 30 l min-1 are more convenient. One tank of liquid nitrogen easilypurges a 10 m3 barrier-film tent to under 1000 ppm oxygen. Using liquid nitro-gen requires no mechanical components and little maintenance, and it is inex-pensive. A considerable amount of ice forms on the top of the tank when a flowrate of 34 l min-1 (49 m3 per day) is maintained for a long period. This may besomewhat inconvenient but it is not a hazard. The tank has a limited shelf lifeof about one month because nitrogen gas is continuously and automaticallyreleased to avoid an excessive pressure increase in the tank at roomtemperature.

21Methods and Materials

Oxygen Scavengers

The availability of a powerful and effective oxygen scavenger, which is basedon the conversion of iron to iron oxide, has been an important factor in thedevelopment of anoxia treatments for pest control. Among a number of compa-nies offering similar products are Atco SA, a French company operating throughdistributors in Europe, and the Mitsubishi Gas Chemical Company, a Japanesefirm that markets its Ageless oxygen absorbers worldwide. Ageless is the morewidely distributed and discussed absorber and will be the basis for most of thediscussion on oxygen scavengers in this book. This material was brought to theattention of Mark Gilberg in 1990, when he was at the Australian Museum, byscientists at the Stored Grain Research Laboratory in Canberra. After someevaluation of Ageless, Gilberg used it in his museum pest control projects andbrought samples to the Getty Conservation Institute, where it was studied foruse in hermetically sealed, inert gas-filled display and storage cases (Lambert,Daniel, and Preusser 1992). A Mitsubishi product bulletin issued in 1994describes the properties, uses, and limitations of this material. Ageless is mar-keted for inclusion in products for retail consumption, primarily food in airtightcontainers, to bring the oxygen concentration below 0.01% and thereby stopoxidative deterioration. The bulletin claims that the low-oxygen atmosphere willeliminate mold, bacteria, and insect infestation and prevent the growth of aero-bic microorganisms. No conditions are given. Small packets of Ageless can befound in sealed containers of common foods like coffee. The widespread use ofthis material in food packages is an indication of its nontoxicity and safety.

The Ageless packet contains a mixture of iron, potassium chloride, water, andsome type of zeolite—one of a class of natural aluminosilicates that selectivelyabsorb water and other small molecules. Oxygen is removed by reacting with ahighly reactive form of metallic iron, which is probably made by the hydrogena-tion of iron oxide. The oxidation is exothermic, and packets can get very warm.Conservators who plan to work with Ageless should experience this thermal ef-fect for themselves by cautiously feeling a fresh packet of Ageless Z-2000 overseveral hours of exposure to air.

Water is required for the reaction of iron with oxygen as seen in the followingequations:

There is generally enough water available in the packets to enable Ageless totake up oxygen in most situations. However, in low humidities moisture willleave the packet, and the reaction will slow; it may essentially stop if the atmo-sphere is too dry. The French manufacturer, Atco, claims that this does not hap-pen with its scavenger product.

Ageless is sold commercially in six forms, which are tailored to specific end uses.The product conventionally used for pest control is type "Z," designed for use inrelative humidities of 50-85%. Ageless Z is supplied in nine different packetsizes, which are graded and labeled to show how many cubic centimeters ofoxygen the packet will absorb. Thus, the largest size, Z-2000, will absorb 2000cm3 or about 16 g of oxygen, which is the amount of oxygen in 10 l of air. TheAtco product is similarly designated; Atco LH3000 will absorb 3000 cm3 ofoxygen. The conservator can determine the number of packets required for atreatment by the oxygen-removal capacity of the scavenger. To maintain theoxygen concentration at a predetermined level with Atco LH or Ageless, it is alsonecessary to measure the rate at which oxygen leaks into a sealed unit filled with

22Chapter 3

an inert gas. This is easily done with a sensitive low-oxygen-concentration ana-lyzer. The number of packets, N, of absorber is then calculated from the equation

where L is the leak rate of the unit (ppm O2/day); V is the volume of the unit (l);C is the oxygen capacity of the absorbent (cm3); and D is the predeterminedlevel of oxygen (ppm).

For example, the number of packets of Atco LH3000 needed to maintain an oxy-gen concentration of 1000 ppm in a 1000 l case leaking 300 ppm oxygen perday would be 8. It is also important to know the oxygen capacity of the scaven-ger when it is to be used by itself to create anoxic conditions—that is, when thetreatment starts in air and does not involve purging with low-oxygen nitrogen.Determining how many packets of what size scavenger are needed to removeoxygen from a flexible bag filled with air is relatively simple. Because oxygenconstitutes 20% of the volume of air, inserting enough packets to equal one-fifth the volume of the pouch will consume all of the oxygen in it. The gross vol-ume of the bag, at least for smaller ones, can be obtained by filling the bag withwater and then measuring the water's volume. Although it may seem logical tothrow in a few extra packets of absorber to ensure a good result, this isn't neces-sary. Ageless Z, as packaged and sold, has as much as a threefold excess capacityfor combining with oxygen. (The exact amount should be verified by using anoxygen monitor.) This information should temper the use of too much material.

The intensity of oxygen removal by a scavenger is another matter. Lambert,Daniel, and Preusser (1992) studied the rate of reaction of oxygen with Ageless.They examined two factors that can affect the time needed to reduce the oxy-gen concentration to anoxic levels: how the packets were arranged in thechamber and the RH. In most experiments the Ageless packets were placed in aholder that allowed the atmosphere access to both sides. When the packets werearranged with one flat side in contact with the floor, the absorption rate slowedby about 20%, suggesting that this type of placement does not critically deterthe effectiveness of treatment. However, operating at 33% RH, which is belowthe recommended range of 50-85%, did decrease the absorption rate sharply.The researchers concluded that Ageless would maintain the oxygen content at avery low level for several years in an inert gas-filled, hermetically sealed displaycase with a moderate leak rate. For purposes of pest control this study providesinformation on effective and economical ways to use Ageless. It also found thatit is easier to remove oxygen at high concentrations than at low levels. The studyfound that the more absorbent used, the more rapidly the oxygen concentra-tions will be brought to kill levels.

It bears repeating that when Ageless reacts with air, in contrast to a nitrogenatmosphere containing oxygen at just a few thousand parts per million, thepackets become quite hot. Thus, when used for oxygen removal from air,Ageless should never be in contact with thermally sensitive objects. Furthermore,air is 20% oxygen, and after Ageless has removed it from a containment bagduring treatment, the flexible bag shrinks in volume by one-fifth. Delicateobjects could be damaged by this partial collapse if they are placed in a tight-fitting plastic bag at the beginning of the anoxia treatment.

The aggressive use of Ageless was demonstrated by conservators at the Austra-lian Museum in Sydney in 1991, when a total of 1,290 objects were treated bysealing them in handmade, oxygen-impermeable, plastic-film bags and remov-ing all of the oxygen solely with Ageless Z-2000. Overall, 1,675 packets of theabsorber were used (Gilberg and Roach 1992). Only about one-fifth of the items


were known or suspected to be actively infested. The cigarette beetle was themost prevalent pest, but the powderpost beetle, the webbing clothes moth, thedrugstore beetle, and the furniture carpet beetle were also identified. Individualitems and groups of objects, supported by matte board when needed, wereplaced in the bags along with the necessary amount of Ageless Z-2000 beforeheat-sealing. The objects consisted of a range of ethnographic materials includ-ing statues, bowls, blankets of bark, baskets, carved wooden panels, spears,headdresses, necklaces, and paintings on bark. Flushing the bags with nitrogenwas not done, as the Ageless served to lower the oxygen in air to anoxic levels.The quantity of oxygen scavenger required was determined from the oxygencontained in a volume calculated from the length and width of the bag and theheight created by the contained object, less the volume of the object. Bagscontaining objects of extreme fragility, such as feathered headdresses, or ofexceptional size were first flushed with nitrogen, and a much smaller amount ofAgeless was used. The bags were made from an ethylene-vinyl acetate copoly-mer, poly(vinylidene chloride), and nylon composite film that has an oxygen per-meability of less than 5 cm3 m-2 per day at 23 °C and 70% RH. This film wasfavored for its low oxygen permeability, good handling properties, and the easewith which it could be sealed using a bar sealer. All sealed bags were placedinside a temperature-controlled cabinet for 21 days at 30 °C. After use, theAgeless was discarded. Examination of infested items a year later failed to findany trace of continuing activity.

Closures

Heat Sealers

The construction of gastight containers using barrier films relies primarily onheat-sealing to make oxygen-impermeable unions. One surface of a heat-sealable barrier film must consist of a polymer, usually either polyethylene orpolypropylene, that has a softening point below 177 °C and is fusible in a man-ner that forms a nonleaking, oxygen-tight seal. Creating the perfect seal requiresa balance of hand pressure, sealer temperature, and press time. Doing this cor-rectly is an acquired skill, but with practice and diligence most conservatorsdevelop a degree of skill that allows them to turn barrier film into perfect anoxiabags easily and competently.

Heat-sealing has long been used in the manufacture of industrial packaging forsuch diverse, atmospherically sensitive materials as food and electronic compo-nents. A wide range of heat-sealing equipment is commercially available.Sources include bagging supply companies, general equipment suppliers, andmore specialized sellers of materials for conservation. Table 3.5 describes severaldifferent sizes and types of heat sealers. The handheld bar sealer is the mostcommonly used tool for conservation purposes. Tacking irons, such as theSealector II or the Seal King, are less frequently used but they are cheaper. Someconservators claim that with experience, they can seal faster with tacking ironsthan with bar sealers, although the tacking irons must be used with a suitablesoft and flat undersurface. A tacking iron's broader seal width—4.45 cm (1¾ in.)compared to 1.43 cm (9/16 in.) for the handheld bar sealer—can be an advantagefor making larger, flexible chambers. The automatic impulse sealer is moreexpensive, but it is cost-effective if there is a large amount of bag sealing to bedone. Spatula and spade-tip tacking irons provide much smaller heating surfacesbut can get into locations where the larger tools may not fit. A small tacking ironis especially suitable when a transparent window is called for in a metallized bag.It can be used to heat-seal the edges of a barrier-film patch to a metallized filminto which a smaller window opening has been cut. Robert Hinerman, workingwith the conservation staff at the Mendicino County Museum in northern Cali-fornia, has fabricated a sealing tool for this purpose. It consists of a metal ring at

24Chapter 3

Type

Table 3.5

Heat sealers.

Bar sealer

Tacking iron

Tacking iron

Automatic impulse sealer

Spatula tacking iron

Spade-tip tacking iron

Description

Handheld1½ lb (0.68 kg)6 in. (15.24 cm) long9/16 in. (1.43 cm) wide100-260 °F(38-127 °C)

One surfacethermostated150-350 °F(65-177 °C)

Heating area:4 in. (10.16 cm) wide1¾ in. (4.45 cm) long150-400 °F(65-205 °C)

24 in. (60.96 cm), 2 jawstabletop modelautomatic time set

Heating area:11/16 in. (2.38 cm) wide¾ in. (1.91 cm) longtemperature to 450 °F (230 °C)

Similar in size to the above

Model

FuturaPortableBarrier Model

Sealector II

Seal King

92-96348

Tacking tool74535A2

Adem C6

Supplier

Packaging Aids

Conservation Materials

McMaster-Carr

National Bag Company

McMaster-Carr

Conservation Materials

Cost

(1996 US$)

140

63

49

637

28

48

the end of an electric heating iron. The ring, which heats to over 177 °C, has anouter diameter of 5.08 cm and a bar width of about 1.27 cm, and convenientlyheat-seals a very functional circular window into the metallized film.

Zippers

The use of plastic zippers, or zip-locks, is another way of sealing a plastic bag.These fasteners are fabricated separately and then heat-sealed into the con-tainer. They appear in two applications: as closures on anoxia bags and asportals to much larger and permanent flexible chambers. Although zippers areconvenient, they leak badly and can be used only on bags for dynamic anoxiatreatments. In this mode, the leakage through the zipper allows the constantflow of humidified nitrogen through the bag. A heavier, better-designed zip-lockis built into large chambers. These have to be made more effective than bagzippers because repeated heat-sealing is not an option for reusable chambers.Even though they are much more gastight than small zippers, these closures arethe part most prone to leakage on large bubbles. However, the application ofgenerous amounts of petroleum grease through the zipper as it is closed hasbeen found to reduce leakage.

Clamps

Clamps for anoxia bags are plastic strips up to 183 cm long and about 1.25 cmwide, generally resembling a horseshoe in cross section. The film layers to bejoined are placed between the two halves of the clamp, which form a tight sealwhen pressed together. There is some disagreement as to the effectiveness of


this closure, but most conservators who have worked with these clamps havefound that they can provide leak-proof closure. In general, however, conserva-tors continue to use heat-sealing. Clamps appear to work better with softerlaminates, such as Marvelseal 360 or Film-O-Rap 7750, than with the lessflexible Filmpack 1193. In Japan the Mitsubishi Gas Chemical Company marketssmall anoxia kits for food preservation consisting of 30.5 cm2 (1 ft2) barrier bags,a small amount of Ageless, a single Ageless-Eye tablet (a passive oxygen moni-tor) for each bag, and a set of plastic clamps, which is some testimony to theutility of this form of closure. These kinds of clamps can be obtained from sev-eral of the companies supplying conservation needs.

Adhesive Strips

Another way to make gastight seals between sheets of barrier film or to providea closure for anoxia pouches is to employ one of several types of sticky tapesproduced by Schnee-Morehead, Inc. These are used industrially to form high-vacuum packaging seals. Oxygen-impermeable closure is obtained by placingthe tape along the inside rim of the bag opening and pressing firmly along thelength of the bag-tape interlay. Another tape, SM 5144, available from Conser-vation Support Systems and other suppliers, is tinted yellow from a pigment thatchanges color when the tape is thermally cured (a process not required for insectanoxia treatment). The tape has a leachable component—liquid polyisobutyl-enes of extremely low volatility—that may be the basis for the gas impermeabil-ity of the seal in much the same way that petroleum grease helps to cut down onleakage through zip-locks on bubbles. John Burke confirmed the effectiveness ofthis tape in studies done at the Oakland Museum. He purchased from an Oak-land hardware store a similar sealing tape that seemed to differ from SM 5144only in its white color. After both tapes were used to seal anoxia pouches, theability of the tapes to make gastight, oxygen-impermeable seals was tested withAgeless-Eye indicators and by leak-testing. Each tape was tested in triplicate.The yellow tape succeeded; the hardware tape failed. A major disadvantage ofthis type of closure, however, is the general difficulty experienced in workingwith a sticky tape. In addition, these materials do not pass the Oddy test, whichdetects the off-gassing of sulfur compounds. Although tape might be conve-nient for those without the tools or experience to carry out heat-sealing, con-servators who construct bags on a routine basis feel that heat-sealing is stillfaster and easier.

Portals

The passage of anoxia gases in and out of plastic containers is done throughnonleaking portals generally made with plastic or brass fittings. The Swagelokbrand is optimal and is available at hardware stores and laboratory supplyhouses. Conservators have found Swagelok O-Seal Straight Thread Male Con-nectors to be the most satisfactory fittings. These are illustrated in catalogswith the designation "[X-Y]-1 -OR," where X is the metal type of the adapter(B for brass, S for stainless steel) and Y is the tubing dimension (400 for¼ in. [6.35 mm], 810 for 1/2 in. [12.7 mm]). The most popular models are the B-400-1 -OR and the B-810-1 -OR. Common laboratory cork borers can be usedto cut precise holes in plastic bags to establish a perfect contact with the O-ring.Bradpoint drill bits should be used to make a clean hole in thicker polymer sec-tions, such as the tough polypropylene caps of humidity bottles. It is helpful towrap the brass threads with Teflon tape to get a tight seal. Figure 3.1 shows across section of a portion of the O-ring sealing area of the Swagelok adapter atthe site where the gastight seal is made. After a circular hole is cut through acontainer wall, the O-ring is positioned against the outside surface; the flange ison the inside against a polyethylene or polypropylene surface. A good seal isobtained by tightening the flange nut. A problem encountered in using these

26Chapter 3

Figure 3.1

Schematic cross section of a portion of theO-ring area of a Swagelok adapter.

metal fittings with barrier films is that the O-ring does not protrude much abovethe surface of the front nut (surface A in Fig. 3.1). This may cause the surface tocontact and rub against the barrier film on extended tightenings, thereby dete-riorating the film. A variety of solutions has been used to solve this problem.One is simply to use an O-ring with a larger cross section. Another is to put athin bed of epoxy resin in the bottom of the circular groove holding the standardO-ring. Alternatively, the outer film surface can be made heavier and stronger inthe area of the fitting by adding layers of adhesive tape.


Surface A

Flange

Pressure

Pressure

Groove

O-ring

Front nut

Outer surface ofbarrier film

Inner surface ofbarrier film

29

Figure 4.1

Three-jar humidification system used with

Vacudyne chamber at the Los Angeles County

Museum of Art. (Photo courtesy of the

Los Angeles County Museum of Art.)

Chapter 4

OperationalProblems andPractices

Humidification

The compressed gases typically used for insect control—nitrogen, argon, andcarbon dioxide—are supplied from commercial cylinders in "bone dry" condi-tion—that is, their water content is in the range of 10-20 ppm. Moisture-containing and moisture-sensitive materials will slowly lose water and undergodimensional and chemical changes if moved from an atmosphere with ambientlevels of water vapor to a much drier one. These materials would include allcellulosics such as wood, cotton, linen, and paper as well as many protein-containing substances such as glue and leather. All of these are materials onwhich insects like to feed. Consequently, to avoid hygrometric shock to museumobjects during fumigation, the gas streams may have to be humidified beforebeing passed into the treatment chamber.

The three-jar humidification system is one method of providing moisture duringtreatments. It was developed by workers at the Getty Conservation Institute andthe J. Paul Getty Museum (Hanlon et al. 1992; Rust and Kennedy 1993; Daniel,Hanlon, and Maekawa 1993) and is pictured in Figure 4.1. As seen in Figure 4.2,which is a diagram of the flow scheme, dry gas from the cylinder is split into twostreams. The first is humidified by bubbling through a column of water held in athick-walled polypropylene bottle of 2-4 l capacity. The stream is now saturatedwith water, although if the flow rate is too high and equilibrium cannot beestablished, it will be less than saturated. The humidified stream then passes intothe second bottle, where it is mixed with the remaining dry stream, which ispassed directly to this bottle. The mixing bottle also serves to knock down anyspray. The combined streams next flow through a third bottle, which holds arelative humidity sensor, and then into the treatment chamber. A system of threeneedle valves, as shown in Figure 4.2, is used to adjust the ratio of flows andthereby the level of humidity in the treatment gas. The pressure within thethree-bottle system is somewhat greater than the ambient pressure. This causesthe gas, typically nitrogen, in the sensor bottle to have a higher RH than theexiting gas filling the anoxia chamber. At 50% RH and room temperature, thedifference is about 10 percentage points—that is, 40% RH in the exiting gaswith a system of the size indicated. One can compensate for this by creating amodestly higher RH in the humidification system than that required in the treat-ment chamber, and monitoring the humidity levels in both locations with a pairof relative humidity sensors. This works satisfactorily if both sensors have a short

Figure 4.2Schematic of three-jar humidification system.

response time and are highly accurate. Such sensors, however, are expensive.Simple closure-type flow regulators, like needle valves, cannot be set to provideflow ratios with precision; without a pair of very good monitors, they coulddeliver a humidity level far from the desired mark. A way around this dilemma isto pass the separate gas streams through relatively inexpensive flow meters asshown in Figure 4.3. This provides an accurate determination of the ratio of wetand dry gas flows and, thus, of the RH entering the chamber. Then, only asingle, simple humidity sensor in the chamber is needed.

Alternatives to the bubbler approach to humidification include placing pieces ofwet cloth in the treatment bags or conditioning the objects at somewhat higherhumidity than that at which they are normally kept prior to treatment. Placinghumidifiers inside the treatment chamber may be called for when working withvery large containment units where the need for delivery of substantial quanti-ties of nitrogen or carbon dioxide in reasonable periods of time would requirehigh inflow rates. This may be difficult to do with a conventionally sized bubblersystem, where troublesome pressure buildups and less-than-saturated streamswould result. At Canada's National Museum of Science and Technology inOttawa, the staff carries out fumigation with carbon dioxide in a custom-builtbubble that measures 6.7 × 2.4 × 2.7 m high. Humidification is done with aBionaire ultrasonic humidifier placed inside the bubble. Distilled or deionizedwater should be used with this unit. Conservators at the Museum of Fine Arts inHouston do their treatments with nitrogen in a 30 m3 bubble. This size requiresthe input of twenty to thirty cylinders of nitrogen per treatment. Humidificationis provided by a 9.46 l (2.5 gal.) steam-type boiler. Two Cole-Parmer minihygro-thermographs are positioned on opposite sides of the interior of the chamber,next to viewing ports. The steam humidifier and a 41 cm (16 in.) fan are posi-

Figure4.3Schematic of new humidification module.

Chapter 4 30

Nitrogensupply

Water Mixing Sensorbottle bottle bottle

RH sensorTo bag

Dry nitrogensupply

Valve foradjustingN2 flow To inlet

flowmeter

Flowmeter

Valve foradjusting RH

Waterbottle

Dropletcollector

Operational Problems and Practices 31

tioned along an interior side wall near the incoming nitrogen so that there isgood mixing of moisture and nitrogen before the humidified gas enters thespace containing the infested objects.

A very popular alternative is simply not to humidify. Many conservators do notrun their equipment with humidification units, and the majority of materialsbeing treated are not subjected to the deliberate introduction of additionalwater. The main factors allowing for this decision are the nature of the anoxictreatment and the objects being treated; whether nitrogen or carbon dioxide isbeing used; and the amount of moisture buffering that occurs. In a dynamic orflowing system, where dry nitrogen continuously passes over objects at a signifi-cant rate, humidification is a must to prevent hygrometric shock. This is lessobvious in a static system where a fixed atmosphere, even one with near zerohumidity, is in equilibrium with items undergoing treatment.

Objects invaded by pests contain primarily cellulosic materials, which have a sig-nificant and reversible water content. Buck (1952) has shown that wood types,including poplar, ash, oak, chestnut, and fig, ranging in age from less than oneyear to over thirty-seven hundred years, closely averaged 10% water content at60% RH; 8% water content at 40% RH; and 6% water content at 20% RH. At22.5 °C, the addition of one gram of water to a cubic meter of air will increasethe RH by 5%. Thus, a kilogram of wood, conditioned to 60% RH, would loseonly 10 of its 100 grams of water to dry nitrogen to bring the entire system toequilibrium at 50% RH. In smaller containments or in holding boxes of card-board or wood or other buffering material, the loss of water from wooden arti-facts could be much less. Kamba (1994), studying how fluctuations in humiditycause dimensional changes in wood kept in a closed case, concluded that woodenclosed with little air and without any moisture-buffering materials is dimen-sionally more stable than wood enclosed with conditioned silica gel. In the caseof textiles, the major humidification concern is with constrained fabrics suddenlybeing exposed to high relative humidities (Michalski 1993). They can shrink dan-gerously in 75-100% RH, especially above 90% RH. Below 75% RH, con-strained textiles show very little tension change. Humidity control is generallynot a concern for unconstrained material such as clothing or carpets. Therefore,with fabrics the main caution for operators of anoxia chambers is that theirhumidification system does not accidentally supply too much water.

Valentín, in 1993, studied the buffering effect of paper conditioned at 40% RHand 80% RH and subsequently maintained in argon atmospheres of 0% RH and50% RH. The results of the four experiments are shown in Figure 4.4. Plasticbags containing the conditioned paper were purged at 2.0-2.5 l min-1 for 4 daysand then sealed and allowed to stand an additional 6 days. The humidified argonmaintained RH equilibrium in the two experiments where it was used. When dryargon was used, the system settled around 65% RH after starting at 80% RH. Inthe plastic bag with paper conditioned to 40% RH, the dry argon actually pickedup enough water to provide a stable RH above 40%. Valentín concluded thatmuseum holdings maintained in modest relative humidity ranges can be easilydisinfested using a nonhumidified inert gas. However, works of art alreadyexposed to high humidities, and sensitive objects such as musical instrumentsor polychromed wood, should be treated with humidified gas.

Valentín found that nitrogen and carbon dioxide behaved similarly to argon.Cipera and Segal (1996) demonstrated this with carbon dioxide on an opera-tional scale. In fumigations in a 30 m3 bubble containing wooden artifacts aswell as textiles packed in cardboard boxes, the RH, after most of the air wasreplaced with dry carbon dioxide, would often drop as much as 10 percentagepoints, but it would return to initial values within forty-eight hours (Fig. 4.5).

Chapter 4 32

Figure 4.5

Hygrothermograph of carbon dioxide fumiga-tion in 30 m3 bubble showing humidity buffer-ing by contents of bubble. (Courtesy of theConservation and Technical Services Division,Canadian Museum of Civilization, Hull,Quebec.)

Figure 4.4a,b

RH change in plastic bags during treatment of paper. Bags were purged with dry argon and humidified argon (50% RH), after starting at (a) high(80%) initial RH, and (b) low (40%) initial RH (Valentín 1993). (Courtesy of Nieves Valentín.)

Conservators using carbon dioxide are generally less likely to use humidificationthan those employing anoxia. Carbon dioxide as a fumigant works most effec-tively at a concentration of 60-70%. The residual air still contains its watervapor; thus, one-third of the original humidity is maintained, and a totally dryenvironment is never present. In addition, the typically large carbon dioxidefumigation unit is usually filled with objects well-wrapped in paper or cotton andpacked into cardboard boxes in which there is significant moisture buffering.Avoiding excess humidification helps to hasten and ensure a complete kill andallays concerns about the formation of carbonic acid.

Days(a)

Days(b)

Days

RH%


Argon versus Nitrogen

While most anoxia treatments are being carried out with nitrogen, the use ofargon deserves consideration. Although helium is also effective, this very lightgas has a relatively high diffusion rate through most plastic films, creating con-tainment problems; it is also expensive. Argon is attractive because its cost inmany locations may be comparable to nitrogen. Additionally, studies by Valentínshowed that the time necessary for 100% assured kill with argon was 65-70%of the time with nitrogen (Fig. 1.3). The species compared at 40% RH, 300 ppmoxygen in argon or nitrogen, and at 20, 30, and 40 °C included the old houseborer and the cigarette, furniture, deathwatch, powderpost, drugstore, book,and black carpet beetles. The study showed, for example, that at 20 °C, it took7 days to kill all life stages of the furniture beetle with nitrogen but only 5 dayswith argon. Argon has been the anoxant of choice for Valentín for pest control inhumid regions where the cost of nitrogen and argon is generally comparable.She found that when using a 6.2 m3 commercial bubble, based on a film madewith poly(vinyl chloride) and reinforced with polyester, the oxygen leak rateafter 11 days was 2.3% with nitrogen and 1.6% with argon. It is not clear whythis should be the case, but the oxygen leak rate was also higher with nitrogenthan with argon for poly(vinylidene chloride) bags—0.6% versus 0.5% overone year. The slightly slower influx of oxygen when argon is used may helpexplain the higher rates of insect kill with this gas.

Koestler and Mathews (1994) conducted a pilot program in 1992 in which theyused argon to treat ancient documents, infested with book lice, from the libraryof the Great Lavra Monastery on Mount Athos in Greece. Seventy manuscriptswere rendered free of lice by sealing them inside plastic bags based on nylon filmas the oxygen barrier. The bags were flushed with humidified argon to bring theoxygen content to 700-1000 ppm before heat-sealing. While Koestler andMathews claimed that 20 days of exposure would have been sufficient in amuseum that maintains a constant room temperature, the variable and some-times quite low temperatures at Mount Athos led to a minimum of 30 days oftreatment. Unfortunately, the pilot program was not expanded to treat theentire, very large manuscript collection, although Rentokil has since done selec-tive fumigations at the monastery with carbon dioxide (Smith 1996).

Whether the cost of argon is comparable to nitrogen or substantially higher maydepend on the location and the availability of suppliers other than prime dis-tributors of research-grade gases. The cost of the two highest purity grades ofnitrogen, based on the February 1996 listed prices of prime suppliers, ranbetween $4.08 and $5.10 per cubic meter in Los Angeles, California, whilethe corresponding grades of argon ranged from $5.10 to $6.80 per cubic meter(Table 3.4). This is based on prices ranging from $37.00 to $65.00 for aT-cylinder of gas. An industrial grade of nitrogen, assayed at 20 ppm oxygen,can often be obtained from secondary suppliers for a fraction of the price of theresearch grade. The savings on industrial argon are less but still significant.

Proponents of argon cite advantages other than speed of kill, but the practicalimportance of these has yet to be demonstrated. For example, argon is some-what heavier than oxygen and less dense than carbon dioxide (Table 4.1). Thismight lead to stratification that would create lower oxygen concentrationsaround insect-infested objects placed on the floor of a containment unit.However, the magnitude of the density difference is not likely to create dramaticstratification effects. Additionally, if there are diurnal temperature swings in thefacilities where the bags are stored, then thermal convection currents within thebags are likely to eliminate any such stratification. Also, there is a perceptionthat microbial or fungal growth on museum objects might feed on nitrogen,which would not occur with argon. However, this is not likely to happen on

Table 4.1

Density of common gases at 1.0 atmosphereand 0°C.

Gas Kg m-3

Nitrogen

Oxygen

Argon

Carbon dioxide

1.25

1.43

1.78

1.98

wooden pieces in the absence of liquid water (Blanchette 1995) and does notpose a genuine threat to the use of nitrogen for anoxia.

Monitors

This section provides a general description of the different types of monitors thatare used with nontoxic gases for pest control. Monitoring equipment for con-ducting treatments can be either purchased directly from manufacturers orobtained from integrated-sensor dealers like Omega Engineering, Inc., the Cole-Parmer Instrument Company, and the Mitchell Instrument Company. (Only afew vendors are mentioned here. A more complete list of suppliers can be foundin Appendix B. Conservators are urged to contact these companies for currentprices as well as more precise information on specifications and availability thancan be provided in this book.)

Electronic Oxygen Monitors

Instruments that are commercially available for measuring oxygen concentrationfall into two classes: trace analyzers and monitors. Trace analyzers determinevery low levels of oxygen with a resolution of 1 ppm and are generally expen-sive. They are used to monitor combustible environments in process industriesand in food-packaging quality-assurance programs, and they are critical foranoxia research.

Less sensitive and less expensive units, called monitors, are suitable for mostwork with anoxia. Instruments with an oxygen concentration range of 0-25%and a resolution of 1000 ppm are available. They are typically used for monitor-ing the safety of work places where oxygen deficiency is a possibility. For pestcontrol, the monitor should have a range of 0-1% with a resolution of 200 ppm.It should be capable of being calibrated to 20.9% oxygen, and it should betemperature-compensated. Monitors can be used in two configurations, eitheras a remote sensor or with a sample aspirator. The remote sensor consists ofprobes, based on oxygen-sensitive, galvanometric membrane electrodes, whichproduce an electrical potential that is a function of the oxygen concentration.This electrical signal is converted to a digital or analog concentration value fordisplay. If the monitor is battery operated, it can be left inside the enclosure, andreadings can be made from the outside through a viewing window. Otherwise,cables need to be passed from inside the enclosure to an external readout unitthrough a hermetically sealed connection. The sample-aspiration configurationuses an internal or external pump to draw a gas sample from the anoxic atmo-sphere for analysis by a sensing cell in the monitor. The sample can be eithervented or returned to the enclosure. Sensor probes deplete as they react withoxygen, losing sensitivity after six months in a normal (air) atmosphere. Tele-dyne Analytical Instruments, Delta F. Corporation, and Gas Tech, Inc., are well-established oxygen-sensor manufacturers in the United States.

Passive Oxygen Monitors

A simple and inexpensive way to determine whether an atmosphere's oxygenconcentration has been reduced to below 1000 ppm is to use a passive oxygen

34Chapter 4

Figure 4.6

Oxygen-related color change in the azine dyemethylene blue.

monitor. This is a ¼ in. (6.35 mm) diameter tablet that is placed inside the treat-ment container; it is blue in the presence of oxygen and red or pink under anoxicconditions. This color change is based on the chemistry of methylene blue, anazine dye employed in the formulation of the monitor (Fig. 4.6). Azine dyes canalternate between an oxidized or a reduced state, and this is indicated by color.Air keeps methylene blue in an oxidized condition with a corresponding bluecolor. In the absence of oxygen and in the presence of mild reducing compoundssuch as ferrous salts or aldehydes, the dye changes to pink.

Scientists at the Mitsubishi Gas Chemical Company have studied the chemistryof this color change and developed the Ageless-Eye oxygen indicator. They haveobtained a variety of Japanese, European, and United States patents coveringproducts marketed under the Ageless-Eye trade name. One of their early formu-lations was based on glucose as the reducing agent in a mixture with triethyl-amine, oleic acid, methylene blue, titanium dioxide, and glycerin (U.S. Patents4,169,811 and 4,349,509). This formulation turned from blue to red at oxygenconcentrations as high as 3.5%, which is much too high for killing insects.There were other problems as well with this early formulation. A relatively highmoisture content was needed for the chemicals to interact and provide a colorchange. In addition, over time, the pink-red color would take on a bluishcast, even in a very low oxygen environment. A more recent patent (Inoue,Hatakeyama, and Yoshino 1994) disclosed that these problems can be overcomeby substituting 1-amino-2, 3-propanediol for the triethylamine. This formulationprovides a color transition that occurs under conditions closer to those neededfor anoxia treatment. It illustrates how dependent accurate monitoring is on thenature and quality of the chemicals comprising the indicator, which may explainwhy conservators have had variable results using Ageless-Eye as an oxygen sen-sor. The composition of Ageless-Eye can be formulated to have the color transi-tion occur at different oxygen levels, and two variations are marketed: type "C"and type "K." Type K changes from blue to pink at an oxygen concentrationbelow 2500-3000 ppm. This is well above the transition zone of type C, which isbetween 1000 ppm and 1500 ppm. The color change appears to occur moreslowly with C than K. Type C operates best at 15-35 °C and 60-90% RH, whiletype K is best at 5-35 °C and 30-92% RH. Type C Ageless-Eye has a deeper bluecolor than type K.

When exposed to an atmosphere containing 5000 ppm or more oxygen,Ageless-Eye is purple or blue. It will turn pink when the oxygen concentrationfalls sufficiently below this level. Quickly reexposing the Ageless-Eye to oxygen,within five to fifteen minutes, brings on the blue color. Going in the other direc-tion, from pink to blue, is slower and more difficult and takes several hours afteranoxic conditions have been reached. The return to pink is brought about by theweak reducing agent contained in the formulated tablet. The chemistry of thereduction requires water, a fact that is reflected in the moisture requirementsfor both types of Ageless-Eye. Gilberg (1989b) reported on the poor indicatorproperties at low RH. He found that at 0% RH, Ageless-Eye failed to turnpink even after prolonged exposure to nitrogen atmospheres of very lowoxygen content.

When Ageless-Eye is used in a treatment, it should be positioned to avoid directexposure to bright light. With time and exposure to light and heat, the capacityof the reducing agent diminishes. This deterioration is generally the cause of


premature failure of the indicator. The manufacturer recommends storing thetablets in a cool, dark, and oxygen-free atmosphere, conditions under which thecapacity of the reducing agent is not used up. This is conveniently done by seal-ing packets of Ageless-Eye in a barrier-film bag containing the Ageless Z oxygenscavenger, and placing the bag in a refrigerator. With careful attention to han-dling and storage, Ageless-Eye has been reused nineteen times over a one-yearperiod (Elert 1996).

Mitsubishi offers Ageless-Eye packaged three ways. Type K comes in single bluepackets that have tiny openings in their windows to allow the treatment atmo-sphere access to the indicator. Type C is similarly packaged in single, colorlessenvelopes, but it also comes in a green-colored strip of ten envelopes. Here, too,the singles have openings to the atmosphere, but the strip sets must be punc-tured. The need for great care in storing Ageless-Eye made conservation supplyhouses reluctant to repackage the manufacturer's standard five-thousand-tabletcontainer into smaller, more affordable units. However, the growing demand forthis indicator appears to have overcome this problem, and inexpensive packagesholding as few as twenty-five type C tablets can now be purchased.

Carbon Dioxide Monitors

A large number of carbon dioxide monitors have become commercially availableto cope with recent stringent requirements for indoor air quality. While thereare systems produced to cover any range of concentrations, environmentalmonitors normally provide measurements over two ranges: 0-2000 ppm and0-5000 ppm. (The normal atmospheric concentration of carbon dioxide isapproximately 350 ppm.) These ranges cover the safety limits set by the U.S.Office of Safety and Health Administration (OSHA). The monitors offer an accu-racy that is 5% or better of full scale, but readings normally drift as temperaturechanges. Calibration must be done periodically and is based on pure nitrogenand a reference mixture of carbon dioxide and nitrogen. The determination ofconcentration is usually based on the absorption of infrared radiation at bandscharacteristic for carbon dioxide. The gas to be analyzed is passed through aninfrared absorption cell operating over a narrow bandwidth where carbondioxide absorbs strongly. Changes in radiation energy are detected, amplified,and sent to the signal-processing portion of the system for display or storage.The monitors contain either a diffusion cell, which provides low-power-consuming stable sensing at a low cost, or a sample aspirator, which has higheraccuracy and long-term stability but at a higher cost.

An alternative to carbon dioxide monitors are oxygen monitors that can be usedto determine concentrations of carbon dioxide in enclosures and chambers.When an enclosure filled with air is flushed with pure carbon dioxide, the oxy-gen concentration decreases. A reduction of oxygen from 20.9% to 10.45% cor-responds to an increase in carbon dioxide concentration from about 0.04% to50%; an oxygen reading of 8% indicates that the carbon dioxide level is at60%, the level that is favored for fumigation. The output from the oxygen moni-tor can be configured to read as carbon dioxide concentration. Bruel and KjaerInstruments Incorporated, Gas Tech, Inc., and Valtronics are established manu-facturers of these sensors.

Relative Humidity Monitors

Many types of RH monitors are available, including indicator strips, hairhygrometers, and electronic instruments based on bulk polymer, capacitance,and chilled-mirror sensors. Most respond well in the range of 20-80% RH. Moresophisticated and expensive instrumentation is needed outside this range. Selec-

Chapter 4 36

tion of a monitor type is based primarily on range, accuracy, response time, size,and cost.

The atmosphere in small sealed bags and pouches, where the change in RH isminimal and occurs slowly, can be monitored using inexpensive indicator stripsor dial-type hygrometers. These can be calibrated at only one RH and tend to beless accurate than electronic sensors, which can be calibrated at both a low and ahigh RH. A more accurate and faster-responding sensor is needed in large cham-bers and bubbles where high nitrogen input, for a variety of reasons, obscuresa precise determination of RH. Among electronic monitors, resistivity bulk-polymer sensors are inexpensive but take up to five minutes to produce 90% oftheir response; monitors based on capacitance generally provide data in lessthan thirty seconds.

RH indicator strips are the cheapest sensors, while hair hygrometers can befound at a variety of prices. Electronic sensors range from a simple, low-cost,bulk-polymer-based indicator, to a highly sophisticated and expensive chilled-mirror dew point sensor. Low-cost electronic RH sensors are available from theHVAC monitoring industry. ACR Systems, Inc., Edge Tech Moisture and Humid-ity Systems, Hy-Cal Engineering, General Eastern Instruments, and Vaisala, Inc.,supply most ranges of RH sensors.

Temperature Sensors

A wide range of inexpensive temperature sensors can be purchased. Theseinclude temperature indicator strips and paints, bimetal and dial thermometers,glass thermometers, thermocouples, thermistors, resistive thermal devices, andinfrared thermometers. All of the above thermometers, except temperatureindicator strips and paints, deliver the accuracy needed for pest control work.Bimetal and dial temperature indicators and glass thermometers are inexpensive.Thermocouples and thermistors are electronic sensors that can be connected tochart recorders or data loggers for automatic monitoring. K-type thermocouplesare the cheapest electronic thermometers. Thermistors are, in general, moreaccurate than thermocouples, but they cost more.

Leak Detectors

Conservators embarking on a program of treatments with barrier-film contain-ers, particularly when the containers are handmade, will find a leak detector tobe a valuable tool. The detection of specific leak points can be accomplished inseveral different ways. It can be done by activating a battery-operated ultrasonicsound generator (35,000-40,000 Hz) placed inside a treatment container, andthen searching for the leak point from the outside with an ultrasonic detector.A detection system based on fluorescence works in a similar way. A tracer gasis released into the pouch (for example, by injection, with a gas syringe), andwhen an ultraviolet light is played over the exterior surface, leaks are detected asa fluorescent glow. The cheapest and most popular device is the halogen leakdetector, which is based on a refrigerant gas such as 1,1,1,3-tetrachloropropyl-ene that is released inside the chamber. The detector nozzle senses changes inthe thermal conductivity of the atmosphere around the leakage point and givesoff a characteristic squawk or squeal.

Monitoring Life Signs

Perhaps the most critical and the most difficult factor to be monitored is theextent and the completion of insect kill. Researchers have investigated a numberof barely measurable manifestations of insect life signs but they have beenunable to transfer their work into a monitoring methodology as satisfactory as


direct examination. This is a tedious process; it requires looking for any evidenceof life after allowing enough time and appropriate conditions for all life cycles tocontinue and reveal themselves, should any life still exist. It has remained for afew landmark studies to make these determinations and establish the conditionswhere 100% mortality is achieved with a high degree of confidence. Valentín,Gilberg, and, particularly, the team of Rust and Kennedy have done this to createTable 2.3, which lists the conditions and minimum times required for 100%assured kills for a large number of the most common museum pests. In the studyby Rust and Kennedy, the adult insects were generally examined for mortalityafter a few days; the other life stages were held for inspection for weeklongperiods up to nine weeks and, in some cases, for as long as four months. Thenumber of insects involved in this study ran into the tens of thousands. With theavailability of information on the minimum number of days required for assuredcomplete kill, conservators have confidently carried out effective fumigations—without concerning themselves with monitoring for life signs—by using muchlonger treatment times, occasionally even extending fumigation by four tofive weeks.

Some interesting approaches to measuring mortality directly have evolved.Koestler (Koestler and Mathews 1994) adapted a procedure developed by theU.S. Department of Agriculture (USDA) to determine insect infestations inwheat. The USDA created a highly sensitive FTIR analysis that could detect anoutput of about one part of carbon dioxide in one million parts of total gas perminute from a single rice weevil in 350 g of red wheat. Koestler was able to usehis adaptation to follow the anoxic termination of insect life in museum objects.A picture frame with suspected infestations was placed in a low-permeabilitypouch, and a steady increase in carbon dioxide was monitored for 5 days. Thebag was then flushed with humidified argon until the oxygen content was below0.1%, and new measurements were made. No carbon dioxide was detectedafter 16 days, and air was returned to the system. There was no additional for-mation of carbon dioxide over 11 days, which was taken to show that 100%mortality had been achieved.

A simpler, less sophisticated device to measure and monitor the respiration of asingle insect was described by Carlson in 1995. The breathing of insects removesoxygen and releases carbon dioxide, but the net effect is a decrease in volume.Carlson designed an apparatus, contained in a constant-temperature bath, inwhich a glass tube holding an insect is connected to a capillary containing a tinydroplet of oil. Movement of the droplet provides a life-sign indication ofmetabolism. With some simple determinations of air pressure, temperature,and movement volume of the drop, it is possible to calculate the number ofmolecules the insect respires. The system is designed for amateur scientists andis inexpensive to put together, but a larger gastight container connected to acapillary could make the unit suitable for determining whether infested materialstill harbors living insects.

Safe Use of Nontoxic Fumigants

Although gases such as nitrogen and argon are not pesticides, there are safetyconcerns about their use. Carbon dioxide is somewhat different. It is not con-ventionally considered a toxic material, but the presence of relatively smallamounts in air causes breathing problems. Local regulations governing the useof carbon dioxide as a fumigant vary and should be ascertained by conservatorsplanning to use this gas for insect control. In Canada, for example, carbon diox-ide is not classified as a fumigant, and users do not have to be licensed. In Cali-fornia and in the United Kingdom, it is listed as a fumigant. In England, Rentokilhas gone through the necessary testing and documentation with the appropriategovernmental authorities and is licensed to do carbon dioxide fumigation. No

Chapter 4 38

one else has done this; consequently, British conservators looking to treatinfested material on-site with a modified atmosphere are likely to use anoxiawith nitrogen or argon instead. The physiological effects of excessive concentra-tions of nitrogen and carbon dioxide differ. At moderate levels of carbon dioxide(e.g., 5000 ppm), they are unpleasant and serve as a warning for exposed per-sonnel to leave the area. Modest oxygen deficiencies (e.g., 12-19% oxygen innitrogen) will cause some individuals to experience a pleasurable sensationunder conditions where unconsciousness can occur without warning. Personssuccumbing to any of these gases need to be moved to fresh air and given resus-citation as quickly as possible.

OSHA literature warns that a drop in oxygen concentration as small as 1.4%, forexample, from the normal 20.9% concentration in air to 19.5%, may start tohave adverse effects on individuals. At 8-10% oxygen in nitrogen, the effectmay be lethal. A general indication of what can happen with either nitrogen orargon that is deficient in oxygen is shown in Table 4.2, which has been providedby the Compressed Gas Association. The indications are for a healthy, averageperson at rest. Factors such as individual health, degree of physical exertion, andhigh altitudes can affect these symptoms and alter the oxygen levels at whichthey occur.

The Compressed Gas Association offers the following suggestions to individualswho may be subjected to oxygen-deficient atmospheres:

1. Never enter a suspected oxygen-deficient atmosphere without properprotective breathing apparatus and attendant support.

2. Analyze the atmosphere to determine if there is a deficiency of oxygen.Continue to monitor during the work process. If the oxygen level is lessthan 19.5%, ventilate to establish good air quality.

3. Be trained on what to expect and how to handle it.4. Positively isolate to a confined area any incoming lines, and ventilate

the area.5. When it is necessary to work in any oxygen-deficient atmosphere, provide

a self-contained breathing apparatus or airline-style breathing mask forall workers.

6. Use an established hazardous-work permit procedure in all confined-spaceactivities.

Table 4.2Safety concerns with oxygen-deficient

atmospheres.

Oxygen content(% by volume)

Effects and symptoms of acute exposure of humans

(at atmospheric pressure)

15-19%

12-15%

10-12%

8-10%

6-8%

4-6%

Decreased ability to perform tasks. May impair coordination and mayinduce early symptoms in persons with heart, lung, or circulatoryproblems.

Breathing rate increases, especially on exertion. Pulse rate up. Impairedcoordination, perception, and judgment.

Breathing increases further in rate and depth, poor coordination andjudgment, lips slightly blue. At this oxygen content or less, anoxia willbring about unconsciousness without warning—so quickly thatindividuals cannot help or protect themselves. Lack of sufficient oxygenmay cause serious injury or death.

Mental failure, fainting, unconsciousness, ashen face, bluish lips, nausea(upset stomach), and vomiting.

8 minutes: may be fatal in 50-100% of cases; 6 minutes: may be fatal in25-50% of cases; 4-5 minutes: recovery with treatment.

Coma in 40 seconds followed by convulsions, breathing failure, death.


A similar tabulation of physiological effects for carbon dioxide is provided inTable 4.3 (Banks and Annis 1990).

These safety factors are usually not an issue with anoxia treatments carriedout with reasonable prudence in barrier bags. They come into play duringoperations involving large units placed in relatively tight quarters where manycylinders of gas or containers of liquid nitrogen may be used. The well-knowntendency of carbon dioxide to accumulate in low-lying and enclosed regions is apotential cause of safety hazards when high-level carbon dioxide atmospheresare used, particularly in leaky enclosures. A number of operators of carbon diox-ide chambers have installed monitors that are activated when levels of this gasrise from the normal 350 ppm in air to 1000 ppm. At this point, either a forcedventilation system is turned on or an alarm goes off. These are not uncommonoccurrences. Because their density is similar to that of air, low-oxygen/high-nitrogen atmospheres are unlikely to present a similar type of hazard. All largechambers require adequate ventilation in the rooms in which they are used toensure that there is no external danger to personnel.

Table 4.3

Safety concerns with elevated carbon dioxideconcentrations in air.

Carbon dioxide content

(% by volume)Effects and symptoms of acute exposure of humans

(at atmospheric pressure)

0.1-1%a

3%

5%

10%

15%

25%

Slight increase in lung ventilation.

100% increase in lung ventilation.

Breathing becomes labored; maximum tolerable concentration forextended periods.

Upper limit of tolerance, with retention of consciousness for a fewminutes.

Unconsciousness occurs within 1-2 minutes.

Rapid unconsciousness leading to death if person not removed withina few hours.

aOSHA time-weighted average of threshold limit is 0.50% or 5000 ppm.

Chapter 4 40

Chapter 5

Anoxia Treatmentin Barrier-Film Bags

This chapter, as well as chapters 6 and 7, presents down-to-earth detail aboutthe work of individuals and institutions leading to the development of proce-dures for insect anoxia. The failures as well as the successes are recounted. Asnobbish conservator might call the tone terribly unsophisticated; this is true andit is intentional. A detailed and unvarnished history of actual experience in a newfield is far more helpful to most people entering it than the usual scientific articlethat glosses over crucial obstacles that may have taken years to overcome. Thegoal of this book is not only to disseminate rigorously proved data but also todescribe many varieties of practice, some perfected only after much experimen-tation and some still needing improvement, so that the conservator consideringanoxia treatment will not have to reinvent the wheel.

General Procedures

The most widely used procedure for the treatment of infested objects is anoxiain barrier-film bags. Quite suitable bags may be purchased from a growing num-ber of suppliers who have been upgrading the quality of their products and areproviding bags based on the latest, most oxygen-impermeable barrier compos-ites available from film fabricators. When these bags are purchased, it is advis-able to obtain a statement of specifications for the film properties, especially foroxygen permeability. Treatment pouches can also be handmade from the samebarrier-film composites that are sold in rolls.

Making bags by hand, instead of purchasing them, may seem like a majorinvestment in time, and perhaps too much trouble. But many conservators, oncethey have learned the procedures and techniques, prefer to buy rolls of film andmake their own bags. The procedure is simple and, with time, so is the execu-tion. The first step is to cut a rectangle of film from a roll that is somewhat largerthan twice the desired size of the bag. The rectangle will be doubled over to pre-pare a pouch of the required dimensions. Care should be taken to cut straightlines to produce a true rectangle so that the edges line up precisely. The filmshould be placed on a flat, horizontal surface with the heat-sealable surface,usually a layer of polypropylene or polyethylene film, face up. The sheet is thendoubled over, and a pouch is created by heat-sealing the two side edges, leavingthe top edge open. Finally, the objects to be treated, along with some type ofoxygen monitor, are placed in the pouch, and the container is thoroughly purgedwith a rapid flow of nitrogen gas introduced via a flexible tube inserted into theopen end of the bag. Then, packets of oxygen absorber are placed in the pouch,and the remaining open end is heat-sealed closed. An important step is leak-testing the bag after the objects have been sealed inside. As described in chapter4, this is done by using a hypodermic syringe to inject into the bag a smallamount (a few cubic centimeters) of detectable gas such as 1,1,1,3-tetrachloro-propylene. The puncture hole is sealed with tape, and the nozzle of a halogenleak detector is run over the bag's exterior.

If an opaque, metallized film is used to make the bag, it is not possible to seeinside to check the performance of the oxygen monitor. The solution to this,when the monitor is a small-diameter indicator tablet, is to install a small win-dow in the opaque film before the open sheet is converted into a bag. The win-dow is prepared from a small piece of clear barrier film; often the plastic wrapperholding the packets of Ageless is used. A 2-3 cm (1 in.) square opening is cutinto the metallized film and covered with a 7-8 cm (3 in.) square of clear barrierfilm with the heat-sealable surfaces in contact. The envelope containing theAgeless-Eye indicator tablet is taped to this window with the envelope's smallopenings facing away from the wall. The window is then heat-sealed to theopaque film with a small tacking iron, with appropriate pressure. The windowis created 5-10 cm from where the final closing seal will be made. This allows

41

the bag to be reused when the sealed end is trimmed off after the treatmentis completed.

Conservators' Experience

John Burke has been a force behind bringing to conservators the message andmethodology of treating museum objects in anoxia bags from the earliest daysof this technology. Burke had worked extensively with barrier films and haddeveloped an expertise in their general use for art conservation. He understoodwell the technical advantages and drawbacks of the different types of polymer-and metallic-film laminate compositions. In a 1992 article in the WAAC (WesternAssociation of Art Conservation) Newsletter, Burke described several simpleexperiments that demonstrated that barrier-film bags based on poly(chlorotri-fluoroethylene) film (Aclar) and containing silica gel as a moisture buffer willmaintain a narrow RH range for many years. Conservation practice at theOakland Museum in California, where Burke worked, had involved the use ofsuch a barrier film to line oak display cases to block fatty acid outgassing, to pro-tect silver objects shipped out for exhibition, to construct low-cost microclimatedisplay cases, and to prepare humidity-controlled storage bags. Thus, Burke waswell positioned to design and construct suitable barrier-film containers for thedisinfestation of museum objects using anoxia, when Gilberg, with whom he andValentín had worked in San Francisco, described the effectiveness and require-ments of this procedure. Burke's earlier studies on the relative performance ofdifferent commercial films as moisture barriers were extended to compare theoxygen transmission rates through the same materials. Initially, he dealt withinstitutions in the San Francisco Bay Area. More recently, in workshops like theone held in early 1996 in San Diego, California, he has been bringing hands-onexperience to a wider audience. At the Oakland Museum, he periodically runsloads of objects through a carbon dioxide treatment in a Rentokil bubble butprefers to use nitrogen anoxia in smaller laminated-film containers. Burke claimsthat nitrogen provides a higher degree of flexibility when working with indi-vidual objects and is faster, cheaper, and safer than carbon dioxide.

Transparent bags can be made using a laminate based on poly(chlorotrifluoro-ethylene), but this material is too expensive for general museum use. A thinlayer of aluminum is less permeable to oxygen and moisture than are organicpolymers, and a composite with a core of aluminum film is relatively cheap.Burke uses aluminized laminates purchased in 3 ft (91.44 cm) wide rolls to makebags up to 3 × 5 ft (91.44 x 152.4 cm) in dimension. Pinholes and cracks in thealuminum film form even in the most carefully crafted containers, and thesesources of leakage can be seen by holding bags up to a strong light and examin-ing the film from the inside out. Burke finds that relatively sizable objects can berid of insects in less time under nitrogen in these bags than by carbon dioxidefumigation in his Rentokil bubble—generally 1-2 weeks in pouches compared to4-5 weeks in the chamber. Typically, a customized bag is constructed to be justlarge enough to hold the infested object along with several packages of oxygenabsorber. After all components are placed in the bag, it is sealed almost closedand flushed with nitrogen several times. The excess gas is then pressed out ofthe bag, and the seal is completed. After several days, the Ageless should bringthe oxygen content down below 0.1%, and the Ageless-Eye indicator will turnpink. The bag stays sealed for several weeks, and the oxygen indicator ischecked periodically to ensure that it is remaining pink and that leakage is notoccurring. After treatment, the seal on one side of the bag is trimmed off, andthe disinfested object is put into appropriate storage. The slightly shortened bagis generally reusable for several more treatments.

Burke favors using a quantity of Ageless several times greater than the calcu-lated required amount. He believes this is cheaper than redoing the treatment if

Chapter 5 42

a small leak occurs. The Ageless packages are reused, along with makeup oxy-gen absorber, in subsequent treatments. He is also redundant in his use ofAgeless-Eye because a spent indicator may give a false reading of leakage. Aftertreatment, both the absorber and the indicators are quickly stored for later use intightly closed, nitrogen-flushed bags. Other conservators have suggested thatthe extended reuse of Ageless and Ageless-Eye should be done only by highlyexperienced workers.

Conservators in northern California quickly accepted this new, safe method forfumigating their objects and eliminating infestations. They were provided withinstructions, demonstrations, and even a popular buying guide to sources andcosts, Materials and Equipment for Anoxic Fumigation, which Burke compiled.An example of how this technology was put in place in a small museum is illus-trated by the experience of the staff of the Mendocino County Museum in Will-its, California. This museum has a fine ethnographic collection and, on occasion,encounters infestation problems. Rebecca Snetselaar, who is both curator andconservator, led a group from Willits on a field trip early in 1994 to the OaklandMuseum conservation laboratories, where they were shown the procedures andequipment that Burke had developed for dealing with infestations. With thenotes from the field trip and the cost data from Burke's guide, they were able toprepare a proposal that enabled them to obtain an Institute of Museum Servicestraining grant to fund an insect control program. Burke conducted a training ses-sion in Oakland for the Mendocino staff and, as a final step, went to Willits tobe sure that all procedures were done properly and effectively. This hands-onassistance by an anoxia specialist given at the conservator's facility appears to beimportant for getting programs into operation at institutions with limited per-sonnel. Nitrogen anoxia for treating infested objects is now used on a routinebasis at the Mendocino County Museum. In November 1995, for example, themuseum received seven Porno baskets on loan from a private collector, which itplanned to use in a special exhibit. Upon examination, one of these basketsshowed signs of infestation. There was a substantial quantity of frass in the bas-ket's tissue-paper wrapping and a 6.35 mm hole in one of the willow foundationrows. The baskets were treated immediately, preventing further damage andprotecting the other pieces in the exhibit.

At the Phoebe Apperson Hearst Museum of Anthropology in Berkeley, Califor-nia, the conservation staff also treats infestations using nitrogen anoxia inbags, but their holdings are much larger than the Mendocino County Museumcollection. Accordingly, there are more treatments carried out, and they are morediverse in nature. There is relatively little ongoing infestation with the internalcollection. Older pieces contain residual pesticides from past treatments, and thisis believed to be a deterrent to reinfestation. The principal source of insect prob-lems is new acquisitions. These are generally held in isolation and periodicallychecked for signs of infestation to determine if treatment is necessary. In anyevent, if the nature of the object causes suspicion, treatment is carried out.Moths and dermestids are most often encountered, particularly where woolens,furs, and feathers are involved. Drugstore beetles are an ongoing, annual sourceof infestation in museum storage. They appear to be extremely selective in theirattack of museum objects but travel widely. Dead adults are found in many stor-age cabinets even though they appear to have done no damage.

Madeleine Fang has been responsible for the conservation of the HearstMuseum collection. She had been using primarily freezing and No-Pest Strips(dichlorvos) to deal with infestation problems, when she became acquaintedwith insect anoxia through contacts with Mark Gilberg and with Thomas Strangof the Canadian Conservation Institute. Fang learned to make and use anoxiabags with John Burke at the Oakland Museum conservation facility. Because ofMarvelseal's low cost, Fang opted to use this aluminized barrier laminate to

Anoxia Treatment in Barrier-Film Bags 43

make bags with a built-in window to watch the Ageless-Eye monitor. However,after a year of dealing with bags that leaked because of poorly sealed windows,she dispensed with the window and the Ageless-Eye and decided to put her trustin well-sealed bags without oxygen monitoring. Now the improved procedureinvolves placing the objects inside a Marvelseal bag, flushing three times withprepurified nitrogen, adding the required amount of Ageless, heat-sealing thebag, and then allowing the bag to stand a minimum of three weeks. Leakageproblems, which have decreased dramatically, are usually evident within a day,as shown by the collapse of the inflated bag. If leakage is detected within a fewhours, the bag can generally be resealed with a portable heat sealer while theobject is still in it. Conservators at the museum have dealt primarily with infesta-tions of carpet beetles and casemaking and webbing clothes moths, but therehas been at least one object infested with drugstore beetles. To date, themuseum has not had a problem with any insects surviving. Freezing and treat-ment with dichlorvos are still done, however. The pesticide is used with severeinfestations, particularly of large objects like a tule boat that was too big to bebagged. The museum owns a Vacudyne chamber designed for use with ethyleneoxide (last used in 1989) and a standard 30 m3 Rentokil bubble, but there hasbeen no need to make these units operational for nitrogen anoxia.

At the Metropolitan Museum of Art in New York, Robert Koestler investigatedboth rigid and flexible reusable chambers and pouches. He found that in termsof ease of bagging, sealing, and maintaining a low-oxygen environment, thepouch was the most successful (1992). The sealing system was a major problemwith both types of chambers. The flexible unit was a minifumigation bubbleoriginally designed for use with methyl bromide. Koestler found that it was notpossible to get the reusable seal to maintain a sufficiently low oxygen concentra-tion for more than a day. The bags he uses, often as large as 1.1 × 2.3 m, aremade from barrier films of low oxygen permeability. After filling a bag withobjects, he purges it with argon to an oxygen concentration of 1000 ppm. Thegas is humidified by equipment similar to the three-jar system described in chap-ter 4. Ageless is added for further oxygen control, and both Ageless-Eye andanother oxygen monitor are used to follow the oxygen level. Koestler empha-sized that bags can be used anywhere and that they are low in cost, easy to setup, and effective. Further, they can be reused until cracks and pinholes form inthe wall, making the bags permeable to oxygen.

The Los Angeles County Museum of Art finds that its collection is most easilycared for by the combined use of a large reusable chamber and plastic pouches.Most of its anoxia treatments are done in a converted 1.3 m3 metal Vacudynechamber, but there is a need for custom-built bags for objects that will not fitinside the chamber—for example, 213.36 × 60.96 cm screens from its Japanesecollection. The museum's bags are constructed of Film-O-Rap. Silica gel condi-tioned to 50% RH is used to control moisture, and the oxygen content is takendown to anoxic levels with packets of Ageless after the bags are purged withnitrogen. Treatments are also done in the transparent bags for those curatorswho prefer not to have their pieces treated in a metal box. These curators areconcerned with the potential for accidents, particularly those that might becaused by humidity levels. They also feel a need to look in on their charges whilethe objects are undergoing disinfestation. While most pouches are heat-sealedclosed, clamps also have been found to provide adequately leak-proof bags. Themuseum has not found Ageless-Eye to be a satisfactory oxygen monitor. Theconservators there prefer to place in the bags a battery-operated, portable Tele-dyne oxygen analyzer that reads over a 0-1% range.

In Europe the leader in using anoxia in bags has been Nieves Valentín at theInstituto del Patrimonio Histórico Español in Madrid. There Valentín has contin-ued the research she started at the Getty Conservation Institute, developing and

Chapter 5 44

widely demonstrating the procedures and requirements for effective anoxictreatments. Valentín's studies of the minimum assured kill times for eight com-mon insect pests using nitrogen and argon at various temperatures (Table 2.3)were all done with plastic bags. Her research results were tested at the archivesof Reine de Galicia in Spain. This historic building has six floors with 9000 mof shelving holding archival material. Approximately 20% of the document col-lections were contaminated by furniture and powderpost beetles. A two-yearprogram using insecticides had been only partially successful in reducing thenumber of adult insects and larvae found in documents. Treatment was switchedto Valentín's procedure, in which infested bundles of documents were placed inlarge oxygen-barrier bags along with Ageless. The bags were purged with argoncontaining 200 ppm oxygen at 30% RH for 10 hours and then sealed and heldat 30 °C for about 4 days. Most recently, this program involved preparing tensuch treatment units each day. On average, the infested bundles containedtwelve adults, twenty-six larvae, and eleven pupae, all of which were eliminatedby anoxia. The procedures used in this work have also been extended to treatthe collections in the archives of Spain's La Coruña.

In the United Kingdom, John Newton of the Research and Development Divisionof Rentokil Division PLC, David Pinniger of Central Science Laboratories, andMadelaine Abey-Koch of the National Trust have demonstrated the constructionand use of a very large barrier bag for anoxic disinfestation. They have provideda well-documented description (1996) of its use in the treatment of a collectionof curtains from the Mottisfont Abbey in Hampshire. The project was under-taken in 1995 both to provide insect control and to evaluate the parameters ofthe treatment. The curtains were commissioned in 1938-39 for the drawingroom of the abbey as part of the decor created by Rex Whistler. The ten cur-tains, each 4.2 m long and 2 m wide, were made of silk trimmed with imitationermine composed of combed white wool and cotton. The silk had deterioratedover the intervening years, and the curtains had been attacked by insects, pri-marily carpet beetles but also clothes moths and fur beetles.

A dual-purpose experimental design was developed that would determine theamount of time required for complete mortality of varied carpet beetles andwebbing and casemaking clothes moths (three species typically found in fabriccollections of this type) while, at the same time, achieving a total disinfestationof the curtains. The fabrics were maintained in nitrogen with less than 4000 ppmoxygen at 50-65% RH and 24 °C. The length of the treatment time was deter-mined by placing larvae of each of the three test species in the bag along withthe curtains and removing samples periodically to evaluate the mortality rate.The monitoring of insect mortality required a bag of unique design. The barrierfilm consisted of a 9 µm aluminum core bonded to a 60 µm layer of low-densitypolyethylene and a 12 µm layer of poly(ethylene terephthalate) to provide sup-port and mechanical strength. A large section of film was folded about the cur-tains, which had been first cleaned, lined with acid-free tissue paper, wrapped incambric, and secured with brass pins. The barrier film was folded so that thepolyethylene layer was on the inside. The resulting unsealed container measured5.7 × 2.5 m and held the curtains near the long fold on the left side. Before thefilm was folded, sampling and gas transit ports were installed in such a way thatthey would be positioned at opposite ends of the right side of the finished bag.The open sides were sealed with an impulse-type heat sealer whose 24 voltheating bar switches off automatically after a few seconds to avoid overheating,as all such sealers do. Continuous seams were made along the margins of thetwo short sides. Further sealing along the remaining edge was done in a mannerthat allowed small cages of insects to be sealed in the bag and exposed to itsatmosphere. Subsequently, new seams were made that allowed samples of theinsects to be removed periodically for examination while maintaining anoxicconditions in the bag.


The treatment was designed to achieve and maintain an oxygen concentrationbelow 4000 ppm. The oxygen concentration within the bag was measured atintervals using a Bedfont Scientific Instrument BT 2000 trace-oxygen analyzerconnected to a sampling port. Similarly, the RH was monitored using a NovasinaMIK 3000 thermohygrometer. On the first day, the bag was taken through fourflushing cycles during which some atmosphere was withdrawn, creating a slightvacuum, and then the bag was reinflated with humidified nitrogen containing nomore than 20 ppm oxygen. This brought the oxygen level to a satisfactory2500 ppm. Two more flushings were made on the second day. Then the cornerson the left side of the bag were cut open, twenty-five packets of Ageless Z-2000and twenty packets of Atco 3000 were inserted, and the unit was resealed. Theoxygen concentration at this point was 5600 ppm. It fell to 1300 ppm after7 days, 300 ppm after 11 days, and 100 ppm after 21 days. It remained at thislevel over a total holding period of 73 days. The RH was maintained at 50-70%,and the temperature in the unheated storeroom late in the year ranged from4 to 20 °C. The insect mortality phase of the study showed 100% kill of all lifestages in less than 12 days. Following treatment, the dead carpet beetle larvae inthe curtains were removed, and the curtains were again cleaned. A week beforethe curtains were rehung, the display room was sprayed with bendiocarbwettable powder, an insecticide effective against carpet beetles; new carpetswere installed; and a daily program of vacuuming was begun. The drapes showno sign of damage or change.

After its work at the Mottisfont Abbey, the Rentokil team went on to build evenlarger barrier-film bags, when it was called on to solve a pest control problem bythe French Ministry of Culture (Smith 1995). The French agency was concernedabout the infestation, primarily by carpet and lyctus beetles, of very large oilpaintings that were distributed among churches in the Provence region of south-ern France. For the treatment, forty paintings were gathered in 1995 andbrought to a heated, secure warehouse in Marseilles. The paintings, some aslarge as 5 × 3 m, were placed on specially designed racks and installed in one ofthree barrier-film bubbles prepared on-site. These containers, each over 88 m3 involume, were the largest ever constructed for treatment by oxygen deprivation.They were made of a laminate composed of polyethylene, aluminum, and poly-(ethylene terephthalate) films, similar to the laminate used to make the anoxiabags for the Mottisfont Abbey curtains. Large sections of the film were cut andformed into a chamber by heat-sealing with a specially designed wide-jawsealer. They were fitted on opposite sides with lead-in ports for gas sampling, aswell as with gas inlets and outlets.

The treatment units were first partially evacuated over a 10-minute period andthen refilled with high-purity nitrogen. The nitrogen was passed through a split-stream humidifier adjusted to provide ambient RH to the gas as it passed intothe bubble. This partial evacuation and purging was repeated several times. Withsuch large bubbles, ten to fifteen flushings may be necessary before a targetoxygen concentration of 2000 ppm can be achieved. The general processingconditions defined for this project were developed from data provided by the1993 Rust and Kennedy study. While the design called for maintaining an oxy-gen concentration below 2000 ppm at 24 °C for 7-14 days, the actual oxygenlevel was well below 1000 ppm all of this time. After five flushings, a Bedford415 trace-oxygen analyzer was connected to one of the gas sampling ports.When the oxygen concentration was found to have fallen below 2000 ppm afteradditional flushings, packets of oxygen absorber were placed in the chamber.These effectively removed much of the small amount of oxygen entering thecontainment by diffusion, pulling the oxygen level down below 1000 ppm with-out difficulty. Levels below 500 ppm were attainable but required that theevacuation-purge cycle be repeated when oxygen levels moved into the 500-1000 ppm region. Generally, it was necessary to do this flushing primarily during

Chapter 5 46

Figure 5.1

Anoxia treatment of large paintings in barrier

bags, conducted in Marseilles in 1995.

the early days of the treatment (Fig. 5.1). All paintings were rendered free ofinsects and were returned to their churches.

Comments

In many of the treatments described in this chapter, and in general, conservatorsoften became involved in unnecessarily long treatment times, complex bag con-structions, and difficult insect examinations. Fortunately, the studies conductedby Rust and Kennedy on minimum assured kill times for the most resistant spe-cies provide the information needed to greatly simplify treatment procedures.The use of metallized films is an option to be used only when cost problems arepressing. Effectiveness of treatment should come first, and the use of clear plas-tic barrier films is recommended.

Time (days) (Average temperature 24 °C)


Each line represents a separate bag

ppm oxygen R = Refill with nitrogen

Selectedoxygen limit

48Anoxia Treatment in Barrier-Film Bags

Chapter 6

Anoxia Treatmentin a DynamicMode

Treating infested objects in a dynamic mode with a flowing stream of nitrogen isconvenient and effective in situations with special needs or unique problems.This treatment approach has been done using small pouches, as well as largerchambers built around sizable or awkwardly proportioned objects. This methodis often routinely carried out with pouches, while larger operations tend to be suigeneris. The dynamic mode has some special features that are characteristic andare seen whenever this form of anoxia is used. When very dry nitrogen passescontinuously over an object, with time and at ambient temperatures, deeply heldwater will be pulled out of the object, and irreversible damage could occur. Equi-librium buffering will not take place, and humidification must be used. Anexception to this may be the treatment of waterlogged objects where both anoxiaand drying may be desired. In addition, oxygen absorbers are not employed inthe dynamic mode; removal of oxygen is achieved entirely by purging.

Treatment in Small Pouches

Lesley Bone, conservator at the Fine Arts Museums of San Francisco, California,has been using the dynamic approach to nitrogen anoxia as her standard treat-ment. Bone became aware of its potential for treating infestation problems fromher association in the early 1990s with Mark Gilberg, who was then working inthe San Francisco area. Anoxic treatments at the Fine Arts Museums began in1995 using reclosable zipper bags based on Aclar. The 1.2 m2 bags were sup-plied by Conservation Support Systems. In a typical setup operation at themuseum, prepurified nitrogen is first passed through a bubbler system to bringthe gas to 50% RH. It then passes into the treatment bag via tubing attached byan O-ring Swagelok fitting. The bag also contains a low-range Teledyne oxygensensor. The fitting is attached to an area of the film surface that has beenstrengthened with overlays of a stout transparent tape. After the infested objectis placed in the bag, the zipper is only partially closed, to allow a rapid nitrogeninflow that brings the oxygen concentration down to 1000 ppm. The zipper isthen sealed as tightly as possible, and the nitrogen flow is slowed to maintainthe oxygen level below 1000 ppm. Runs of 2 weeks have been sufficient to killspecies endemic to the museum's collection, namely powderpost beetles andwebbing clothes moths. Incoming wooden ethnographic materials—forexample, African carvings—may contain termites and are routinely treated inthe same manner. Bone has made bags by hand from Marvelseal film. But shefinds this to be more work than using zippered bags, which can be reused aboutthree or four times.

David Casebolt, museum specialist in conservation with the San Francisco Mari-time National Park, has used a variety of procedures to keep the park's archivalcollection free of insects. These include anoxia in refitted and gasketed alumi-num containers from navy surplus, as well as oxygen deprivation with Agelessalone when treating small objects in handmade pouches. His facility is heavilyinvolved with archival material that is most frequently infested with silverfish.The powderpost beetle, beetles in the Dermestidae family, and termites areother species often encountered. Casebolt uses nitrogen anoxia in the dynamicmode for more sizable objects, which go into larger bags. A major area of acqui-sition for the maritime complex is archival collections, which are often receivedin a damp condition. A bubbler system, which would normally be used in thedynamic mode, is usually not employed with these collections. In this way, theanoxia treatment is also used to dry the collection.

Christoph Reichmuth has used the dynamic method, which he calls "nitrogenflow fumigation," for projects ranging from studies of the effect of temperatureon kill time to treatments of wooden sculpture and other objects. The mortalityrate studies were directed at webbing clothes moths, subterranean termites, anda variety of beetles that are common museum pests (Reichmuth et al. 1993;

49

Unger, Linger, and Reichmuth 1992). For these studies, infested objects as wellas control insect specimens were placed in polyethylene bags that were purgedone to three times with nitrogen. The pouches were inflated and kept at a 1%oxygen level with a slow nitrogen inflow at 5-10 Pa. The RH was held at 55-60% with aqueous solutions of glucose or calcium nitrate. The minimum killtimes recorded by Reichmuth were substantially longer than those found byValentín or Rust and Kennedy. However, he had used a higher oxygen level andlower temperatures than the other investigators. For example, Reichmuth deter-mined a kill time of over 500 hours for the powderpost beetle at 25 °C and 1%oxygen, while Rust and Kennedy indicated that only 120 hours are needed at0.1% oxygen at the same temperature. For the furniture beetle, Valentín foundthat 168 hours were needed to eliminate all life stages at 30 °C with 0.1% oxy-gen. Reichmuth required 500 hours at 35 °C and 1% oxygen; dropping the tem-perature from 35 to 16 °C led to an increase in treatment time of 2 weeks, for atotal treatment time of 836 hours. Quite clearly, higher oxygen concentrationsare responsible for extremely long treatment times. Additionally, as has nowbeen thoroughly established, lower temperatures extend the treatment periodeven further.

Reichmuth describes several other successful treatments achieved with nitrogenflow fumigation, albeit with excessively long exposure times at the high oxygenlevel he uses. Two sculptures and a painting—all large, wooden pieces infestedwith furniture beetles—were rid of pests by runs conducted at 25 °C for5 weeks. The painting, by Lucas Cranach the Younger, is a portrayal of theLast Supper on a 2.1 × 2.6 m wooden table. One of the treated sculptures wasa polychrome altar, a shrine to the crowning of Mary. The altar holds threefigures, each 1.1m high. The other sculpture was a Pietà, without mounting,made of limewood and measuring 85 × 55 cm. For these treatments, controlsamples of all life stages of the furniture beetle—some encased in wood toexamine the diffusion of nitrogen into this medium—were added to the con-tainer along with the infested object. The control insects served as monitors ofthe treatment's effectiveness. Reichmuth noted that the disadvantages of nitro-gen fumigation were the long treatment time at 1% oxygen and the lack of aresidual protective effect.

Treatment in Constructed Containments

Objects at the J. Paul Getty Museum

In 1988, when Gordon Hanlon came to the J. Paul Getty Museum in Malibu,California, to work as a conservator in the Decorative Arts Conservation Depart-ment, he found that problems with insect infestation occurred infrequently. Newacquisitions, primarily furniture from Europe, were fumigated and made free ofinsects close to their point of origin prior to shipment. When an in-house needfor disinfestation arose, objects were treated with methyl bromide at a fumiga-tion facility about thirty miles away. However, a variety of concerns abouttransportation, security, safety, and cost of this procedure were raised, making itdesirable to find a procedure that could be done safely at the museum. Theseconcerns came to a focus in 1991 when a 220-year-old French fire screen madeof walnut was found to have a hole penetrating the wood with a small amountof frass beneath the hole. A pile of eggs was also uncovered, evidence of aninfestation by furniture beetles. This event occurred at about the time that work-ers at the Getty Conservation Institute were demonstrating the effectiveness ofnitrogen anoxia, and Rust and Kennedy were gathering data on the conditionsgenerally needed to kill all life stages of the most common museum pests.Hanlon decided to try nitrogen anoxia on the fire screen. He believed that hecould maintain the required treatment conditions by building a container aroundthe object and purging the system with humidified nitrogen. An Aclar pouch was

50Chapter 6

Figure 6.1

Italian armchair under treatment.

custom-tailored into a tapered, almost triangular, shape to fit the fire screen. Theconstruction, done with a small, hand heat sealer, took only about two hours,although Hanlon believes that a larger bar sealer of the type used in paperconservation would have shortened the construction time. Dynamic nitrogenanoxia, as previously noted, requires humidification. This was done using thesplit-stream humidifier described in chapter 4 as well as Arten gel as a humiditybuffer. The system was first flushed rapidly with nitrogen containing only 5 ppmoxygen until the oxygen level in the container fell to 1000 ppm. The nitrogenflow was then decreased and adjusted to maintain the oxygen concentration at1000 ppm, and this was monitored by aTeledyne model 316 trace-oxygen ana-lyzer over a two-week period.

Following the successful elimination of insects from the fire screen, the Aclar bagwas reused to treat a small French console table that was infested with wood-boring insects. The third object in this early series of treatments was substantiallylarger, a 260-year-old Italian armchair that was infested with furniture beetles.A 1.5 m3 container, again of Aclar laminate, was constructed to fit around thechair, which was placed on a table holding much of the monitoring equipmentand the three-bottle split-stream humidifier (Fig. 6.1). The temperature and RHwere measured with a Vaisala HMP 133Y monitor. The oxygen level was fol-lowed with two instruments: a GC Industries oxygen monitor, model GS502,with a 33-475 sensor that provides data down to 1000 ppm; and the Teledynemodel 316 analyzer, which can determine oxygen levels to a few parts per mil-lion. Data were collected by a Campbell Scientific 21X data logger and stored.The careful, continuous recording of oxygen concentrations in this study pro-vided insight into the mechanics of the gas flow needed to achieve and maintainoxygen levels below 1000 ppm during a dynamic run. In a preliminary experi-ment to determine the rate of oxygen leakage into the system, the bag holdingthe chair was first purged with nitrogen containing 5 ppm oxygen at 55% RHat a high input rate of 7 l min-1, while temperature, oxygen concentration, andRH were monitored. This lowered the oxygen concentration to 400 ppm in40 hours. At that point, the inlet and outlet valves were closed, and a slightbulging of the treatment container indicated that the pressure inside was some-what higher than the pressure outside. Leakage began to equalize the pressuresover 24 hours. This corresponded to a slow increase in oxygen concentrationover this period, followed by a faster increase over the ensuing 72 hours until

51Anoxia Treatment in a Dynamic Mode

Figure 6.2

Determination of oxygen leakage rate duringtreatment of Italian armchair.

Oxygenppm

Time, hours

the pressures roughly equalized (Fig. 6.2). During this preliminary experiment,the oxygen leak rate was determined to be 490 ppm (735 ml) per day.

During the actual treatment of the chair, the bag was initially purged with nitro-gen at 7 l min-1 for 60 hours, and this brought the oxygen concentration to800 ppm (Fig. 6.3). The nitrogen flow was then decreased to 0.5 l min-1, and theexit valve was partly closed. The oxygen concentration remained relatively con-stant, increasing to only 900 ppm after an additional 120 hours. This demon-strated that a low flow rate of 720 l per day into a 1500 l bag can compensatefor the leakage of 735 ml of oxygen per day into the bag. It was necessary tochange the nitrogen cylinder at this point, and the flow was increased to2.0 l min-1, where it was maintained for the rest of the 2-week treatment. Thishigher rate brought the oxygen down to 400 ppm, a level where anoxia is evenmore effective than the recommended 1000 ppm for dispatching insects. Failureto find any sign of insect life after four years posttreatment is proof of the suc-cessful disinfestation of the fire screen, the wooden table, and the armchair. Thisled Hanlon to conclude that a continuous nitrogen flow into a large Aclar plasticbag can maintain an oxygen concentration well below 1000 ppm for the dura-tion of treatment and provide a highly effective procedure for pest control.

Figure 6.3

Oxygen concentration in treatment bag

during disinfestation procedure.

Oxygenppm

Time, hours

Chapter 6 52

7.0 liters per minute

0.5 liters per minute

2.0 litersper minute

Back Seat Dodge ’38

Steven Colton, conservator at the Los Angeles County Museum of Art (LACMA),realized that he had a major infestation problem one day early in 1993 when hefound six webbing clothes moths during a routine inspection of one of the muse-um's most popular exhibits, Edward Kienholz's Back Seat Dodge '38. Five juve-nile insects were inside the car, eating floor carpeting, and a sixth moth flew outof the trunk when it was opened. The contemporary sculpture was controversialwhen it was acquired in 1966 because one door was open, revealing a sculpturalportrayal of a drunken, life-size couple making love in the rear of the car. LosAngeles County supervisors denounced the work as revolting and blasphemous,and pressed the museum to remove it. The trustees unanimously rejected theirappeal as an infringement on the museum's obligation to present works thatrepresent an honest statement by a serious artist. The sculpture remained,although for a while, the car door was closed and a museum guard was postedto prevent viewers from peeking inside. Twenty-seven years after the uproarthat rocked the art community, Back Seat Dodge '38 suffered another attack.But, as described by the Los Angeles Times, "this time, the assailants weren'ttwo footed politicians, they were winged moths and when they got a load of theDodge, the wayward insects didn't see obscenity, they saw home and dinner"(Muchnic 1993).

The options for dealing with the problem at the time were limited. The sculpturewas a complex piece containing a wide range of materials: leather, paper, cloth,metals, synthetic resins, glass bottles, flocking material, and a variety of adhe-sives. The museum was loath to send it off-site for pesticide treatment becausetransportation would risk damage to the fragile, deliberately dilapidated art-work, and conventional fumigation chemicals might harm the materials. The carwas too large to fit into a pouch, and there were no big bubbles or chambersavailable similar to ones that others have found suitable for the anoxic treatmentof large objects. LACMA turned to the Getty Conservation Institute for help. Themuseum's timing was good: GCI personnel Shin Maekawa and Vinod Daniel,working with Gordon Hanlon of the J. Paul Getty Museum, had just gonethrough the process of learning how to use barrier film to build containmentsabout large, awkwardly sized objects for dynamic anoxia treatment. At the sametime, a GCI-sponsored program at the University of California, Riverside, underthe guidance of Michael Rust, was collecting the data needed to deal with web-bing clothes moths. The studies by Rust and Kennedy found that holding thisspecies of moth for 96 hours at 25 °C in nitrogen containing 1000 ppm or lessoxygen was sufficient to create confidence that all life stages were exterminated.These data defined the conditions for the successful anoxia treatment of theBack Seat Dodge '38 carried out by the LACMA and Getty teams.

The gallery of the Anderson Building, where the sculpture had been on dis-play, was closed for ten days while the container was built and the treatmentcarried out. The car, on casters, was rolled onto a sheet of plywood placed over alarger sheet of transparent laminate film based on Aclar as the oxygen barrier(Fig. 6.4). Additional sections of this film were cut, assembled, and heat-sealedto the bottom film layer, creating a tight-fitting container completely enclosingthe sculpture. At its maximum dimension, the container measured 6.1 × 3.7 ×1.7 m. It had a surface area of 23 m2, and construction required 44 linear metersof heat-sealing. The estimated volume of the bag was 5.5 m3. However, voidspaces such as the interior of the car and trunk, the area under the hood, andother spaces were filled with nitrogen-charged balloons to reduce the chambervolume to 4.7 m3. This is now considered unnecessary. The oxygen leak ratethrough the bag was 167 ppm per day during the initial purge. Two industrial-grade nitrogen streams, each flowing at 5400 l h-1, were passed first throughseparate split-stream bubblers, then into the bag, and out an exit duct. As the


Figure 6.4

Kienholz's Back Seat Dodge '38 undertreatment. (Photo courtesy of the Los AngelesCounty Museum of Art.)

oxygen level dropped, the industrial grade was replaced with prepurified nitro-gen. Continued rapid purging with nitrogen containing 5 ppm oxygen at 45%RH quickly brought the oxygen level in the bag to below 1000 ppm, where itwas maintained for 8 days under a continuous gas flow at 30 l h-1. Temperature,RH, and oxygen concentration were monitored continually. After treatment,dead larvae and adult insects were found in the car. When the project was com-pleted, the equipment was dismantled, and the car was rolled out of the bag andput back on display.

The Spanish Piano

Early in 1992 inspection of a historic grand piano belonging to the FundaciónBartolomé Pons Cabrera in Valencia, Spain (Fig. 6.5), disclosed that the pianowas host to a major infestation of furniture beetles, both adults and larvae, andlesser amounts of deathwatch beetles. The piano was treated by Nieves Valentín

Figure 6.5Bartolomé March piano being prepared fortreatment. (Photo courtesy of NievesValentín.)

Chapter 6 54

of the Institute de Conservación y Restauración de Bienes Culturales (now calledthe Instituto del Patrimonio Histórico Español), Madrid. The first step was tobuild a large 6.3 m3 tent around it from Saranex film (a laminate of polyethylene,poly[vinylidene dichloride], and ethylene-vinyl acetate copolymer made by DowPlastics). To start the treatment, the tent was partly evacuated and refilled twicewith high-purity argon to quickly lower the oxygen content. The argon, condi-tioned to 40% RH, was allowed to flow through the chamber for 8 days. Theoxygen level fell to 100 ppm during this period, while the temperature wasmaintained at 24-25 °C. After 8 days, thirty packets of Ageless Z-2000 were putinto the bag, the argon flow was terminated, and the unit was sealed so that itoperated in a static mode for an additional 6 days. The effectiveness of thetreatment was demonstrated by the complete mortality of the larvae of longhornborer beetles that were placed in the system as monitors.

Anoxia Treatment in a Dynamic Mode 55

The Getty Barrier-Film Tent

The craftsmanship required to make pouches can be applied to fabricatingreusable containments of any size. Maekawa and Elert (1996), members of theScientific Program at the Getty Conservation Institute, have described the con-struction of a prototype 10 m3 barrier-film chamber that they prefer to call a"tent." This tent, with accessories and a large-capacity nitrogen source, offers asystem that provides convenient and effective treatment (Fig. 7.1). After purg-ing, it will maintain a nitrogen atmosphere containing less than 3000 ppm oxy-gen at ambient temperature and at suitable moisture levels for a month.

A key parameter of this system is the oxygen permeation rate of the tent wall.This is important because it determines the amount of nitrogen needed over thetreatment period to keep the oxygen level below critical concentrations. Theauthors have found that wall material based on poly(chlorotrifluoroethylene) oraluminum film is not suitable. Oxygen transmission rates of laminates using thefluoropolymer were too high (Tables 3.1 and 3.2), while films based on alumi-num foil required great care to avoid forming pinholes on repeated use. Filmpak1193, a strong laminate of somewhat limited clarity with an oxygen permeationrate of only 0.1 cm3 m-2 per day, was chosen. A schematic of the tent is shownin Figure 7.2. The film was purchased as a 1 m wide roll. Sections of film wereheat-sealed to form panels large enough to construct a tent measuring 2 m oneach side and 2.5 m high. Sealing was done with a tacking iron, approximately2 cm wide, instead of a bar sealer.

Stout loops of film were formed along two parallel top edges to attach a liftingdevice that made it possible to raise the tent for easy placement of objects. Thetent was designed to fit over a support frame made of aluminum tubing. Sealingwas done by extending the film outward from the bottom of the tent to create a30 cm flange that pressed against a flat base. The base could be as simple as asmooth concrete floor coated with several layers of paint or a large flat sheet ofplastic. To prepare a seal that would maintain the oxygen level below 3000 ppmfor the duration of treatment, two lightly greased large square gaskets, preparedfrom self-adhesive rubber strips, were first placed in parallel on the flangearea of the base. Then the tent was lowered, and its flange was adjusted to bewrinkle free. Lengths of cast acrylic, 1 cm thick and 15 cm wide, were placed onthe flange perimeter, and these were weighted with bags of lead shot to providea hermetic seal (Fig. 7.3). Over a one-month period, an oxygen leak rate of50 ppm per day was experienced. This made it possible to keep the oxygenconcentration well below 3000 ppm for more than four weeks without furtherpurging, after starting at 1000 ppm. An adequate seal could not be obtained bytaping the flange to the floor, despite many attempts and variations.

Reusable AnoxiaSystems

Figure 7.1

Schematic of anoxia treatment system.

57

Chapter 7 Flexible Chambers

Filmpak1193tentDry nitrogen

supply

Inletshut-offvalve

RH Inletconditioning flowmeter

module

RHoxygensensor

Datalogger

Outlet shut-offvalve Nitrogen

outlet

Outletflowmeter

Pump

Large systems require substantial amounts of nitrogen during the initial purge,and this can be delivered from a bank of cylinders hooked up to a manifold.When especially large quantities of gas are required, an easier and cheaperapproach is to use a tank of liquid nitrogen or the Prism nitrogen generatoroffered by the Air Products and Chemicals Corporation (chapter 3), which cansupply 41 m3 of gas per day containing less than 100 ppm oxygen. The chamberalso requires the type of RH conditioning module described in chapter 4 andshown in Figure 4.1. Equivalent throughputs in the flowmeters would deliver anRH of 50% if the nitrogen bubbling through the water became saturated. Athigh gas flow rates this does not happen, and it is necessary to calibrate flowrates against measurements of RH inside the tent.

The treatment system, including the tent, was constructed in sixteen hours andcost US$2,500 for materials. The cost could be reduced by one-third by usingpoly(vinyl chloride) tubing instead of aluminum for the frame, and sand insteadof lead for the weights. The leak rate of the 10 m3 unit was only 50 ppm(500 ml) oxygen per day. This allowed the system to be purged to 1000 ppmoxygen and then left in a static mode without any oxygen absorber for morethan 30 days before the oxygen level would rise to about 2500 ppm. The

Figure 7.3

Schematic of hermetic seal.

Chapter 7 58

Figure 7.2

Schematic of anoxia tent.

Rope (lifting device)

Aluminum frame(lifting device)

Nitrogen out

Tent coverFilmpak 1193

Acrylic plates

Rubber strip

Acrylic sheet

Nitrogen in

Filmpak 1193

Self-adhesive rubber strips

Lead weight

Acrylic plate

15cm

1 cm

development of this flexible and reusable chamber resulted in a superior system.The prototype, however, was not used for any experimental study or for thetreatment of insects. The goal of the Getty project was to demonstrate theproduction and testing of an anoxia chamber that could be built in a museumconservation facility.

The Small Rentokil Bubble

In 1991 Rentokil offered a 6 m3 off-the-shelf portable and reusable fumigationsystem for use with carbon dioxide. In 1997 Elert and Maekawa evaluated anupdated model of the small Rentokil bubble for its ability to kill insects by nitro-gen anoxia. This unit was similar in appearance to the earlier version, but thepolymer films forming the flexible walls had been changed to provide lower oxy-gen permeability. The bubble measured 1.8 m2 at the base and 1.7m high andhad an overall surface area of 19 m2. Two ports provided for gas passage, andhermetically sealed connectors were used for sensor signal lines. The shape wassustained by internal pressure instead of a frame. Satisfactory sealing of a well-greased zip-lock-type closure was confirmed by leak detection procedures. Thebubble was easily purged to 1000 ppm oxygen using liquid nitrogen as thesource of the nitrogen gas. A Teledyne trace-oxygen analyzer, model 317, wasused in a series of tests to determine the oxygen permeation rate. Over one- totwo-week test periods, the average oxygen leak rate was found to be 450 ppmper day. This is close to the value calculated from the oxygen transmission ratesof the barrier films used to fabricate the bubble. Only 1.9 l min-1 of nitrogen isrequired to maintain the oxygen concentration below 1000 ppm, and this makesthe bubble very effective for anoxia. The evaluation of this unit did not extendto using it for studies with insects.

Dale Kronkright, senior conservator at the Museum of International Folk Art,part of the Museum of New Mexico (MNM) in Santa Fe, has adapted the6 m3 Rentokil bubble to treat the museum's collection with nitrogen anoxia.Kronkright chose this approach after an encounter with a severe infestationshortly after joining the MNM in November 1993. Clothing moths had attackeda collection of Turkish folk art containing a lot of woolen textiles, and this wasdealt with by freezing all of the textiles. Fortunately, all of the objects were inone gallery, and the infestation was contained, but the episode was sobering. Itforced the museum to close the show for three weeks, which was a severe dis-ruption. The museum considered alternative procedures for future infestationproblems and decided to order a chamber for anoxia treatments.

In September 1994, Kronkright negotiated with Power Plastics, the companythat manufactures the bubbles for Rentokil, to prepare a flexible chamber tomeet the museum's needs. He wanted a moderately sized unit that would oper-ate effectively with nitrogen instead of carbon dioxide. He chose nitrogenbecause he believed it to be more efficient and faster than carbon dioxide. Speedof treatment was important because of the demands of the museum's programschedule. A bubble—essentially a 1.8 m cube—was constructed. The tent col-lapses to fit into a duffel bag, 0.6 m in diameter and 0.9 m long, and has beenset up in a number of different galleries, collections areas, and loading docks.Infested objects, either boxed, supported in packing, or stacked on portableshelving or roll carts, are moved into the bubble through a front door. The tent isabout one-fifth the size of the Rentokil bubble conventionally used with carbondioxide. The size was selected so that the tent could be easily moved to any ofMNM's four component museums, which are at separate locations in Santa Fe.This downsizing also substantially reduces the amount of nitrogen needed topurge or fill the unit. The need for nitrogen was further reduced by constructingthe chamber walls of a composite film based on a chloropolymer. This has a

59Reusable Anoxia Systems

lower oxygen permeability than Rentokil's larger 30 m3 chambers, which aredesigned for use with carbon dioxide and built of reinforced polyester.

Once the unit was delivered, the staff had to do a great deal of adjusting andmanipulation of exhaust valves and seals to reduce leakage to manageable levelsso that runs could be made without the excessive use of nitrogen. A documentpocket with a large hole had to be heat-sealed closed. The triple zip-lock sealleaked continuously and had to be caulked. Packing the main seal with house-hold petroleum jelly also reduced the rate of oxygen permeability.

The unit was used seven times in 1995. The chamber was initially fitted with anadjustable internal framework of PVC pipe. Tetrachloroethylene is used as theleak-detecting vapor with a TIF Instruments model 8800 gas detector. Thebubble is inflated to a positive pressure with 99.95% nitrogen until the walls andceiling of the bubble are distended. The gas detector is then used to find leaksalong all closures, valve fittings, and seams. Buckled or improperly seated areasof the zip-lock seal are easily found using this method. After a proper seal hasbeen ensured, the chamber is flushed with nitrogen until the oxygen level is2000 ppm or less. One T-cylinder of nitrogen lowers the oxygen concentrationfrom 20% to 3%. The level decreases from 3% to 1% with a second tank, from1% to 6000 ppm with a third tank, and from 6000 ppm to 2000 ppm with anadditional half tank. A Cole-Parmer 32014-13 direct-reading flowmeter isattached to the exit regulator and allows the flow of nitrogen to be reduced tothe leak rate of the chamber, which is about 100 ppm in 24 hours. Kronkrighthas found that the lowest rate of oxygen diffusion into this bubble attained sofar is about 500 ppm in 24 hours. With the slow dynamic flow of nitrogen madepossible by the addition of the flow meter to the exit regulator, the MNM bubblecan maintain an oxygen level of 2000-3000 ppm for a 72-hour period. Whenthe oxygen exceeds that level, the bubble is flushed again to bring the level backdown to 2000 ppm. Treatment runs last 10 days, and this requires the bubble tobe repurged twice during the run. Each run requires six cylinders of nitrogen.When needed, the chamber is humidified with the conventional three-bottlebubbler system described in chapter 4.

The bubble was used to eradicate a number of species that had afflicted themuseum's collections, including casemaking and webbing clothes moths as wellas carpet, furniture, varied carpet, and odd (Thylodrias contractus) beetles.Now, objects showing signs of infestation, and all new acquisitions that are inany way suspect of carrying insects after a one-month quarantine, are treated.The museum has never found any species to survive ten days of anoxia.

The Large Rentokil Bubble

Rentokil's standard large plastic chamber, known at one time as the "B & G MiniBubble," was designed and widely used for fumigation with toxic gases such asmethyl bromide. Since the late 1980s, this container has also found favor withmuseums for carbon dioxide fumigation. However, it has generally not beenconsidered practical for nitrogen anoxia because of oxygen leakage problemsthat make reaching levels below 1000 ppm daunting, and the large number ofnitrogen cylinders required per treatment.

Steven Pine, conservator at the Museum of Fine Arts in Houston, Texas, neededa large-volume fumigation system that he preferred to use with nitrogen ratherthan carbon dioxide (Pine, Daniel, and Maekawa 1993). He based his preferenceon data indicating that minimum assured kill times were better defined andshorter with nitrogen than with carbon dioxide, and on the feeling that nitrogenwas more inert and safer. Despite the Rentokil bubble's apparent unsuitability fornitrogen anoxia, Pine decided to try to adapt it, and in 1991 a 30 m3 bubble was

60Chapter 7

purchased by the museum. This standard unit has a base 3.5 m square and aheight adjustable to a maximum of 2.4 m. Entry to the bubble is gained througha horizontal zip-lock seal above floor level. The unit is equipped with three portsfitted with quick-release hose connections that have internal check valves thatclose under positive pressure from the interior. They are used for controlling theatmosphere within the bubble. Additional ports for connections to internalhumidity, temperature, and oxygen monitors as well as for electrical extensioncords were patched through the chamber wall using two-piece poly(vinyl chlo-ride) couplings, rubber gaskets, and refrigeration putty for sealing. The Houstonbubble was installed on a smooth, level surface, and cardboard was placed overthe interior floor to protect the bottom plastic from puncture. The conversionfrom air to a low-oxygen atmosphere is facilitated by partially evacuating thebubble with a small vacuum pump before passing in high-purity nitrogen. Aframe of poly(vinyl chloride) pipe was constructed inside the chamber to protectthe contents from walls that might collapse during this partial evacuation. Typi-cally, after the chamber is loaded with materials to be treated, cardboard isplaced between the objects and the walls for protection. A clear film windowprovides a good view of the oxygen analyzer, hygrothermographs, and theobjects behind them in the bubble.

Humidification is provided by an interior steam evaporator placed so that a16 in. (40.64 cm) nonoscillating fan sends the vapor stream against the incom-ing nitrogen port and along the walls rather than directly at the objects. In trialruns, when the steam evaporator was not used, the RH fell to a level below thehygrothermograph's ability to measure—that is, below 10%. During treatment,an RH of 57% is maintained with the steam evaporator set at maximum, the fanon high, and the nitrogen at a delivery pressure of 414 kPa.

Before the bubble was put into routine operation, two steps were taken to ensurethat the enclosure would hold oxygen below 1000 ppm for the time required fortreatment. The first was to find and eliminate any leakage at the seams or theplastic zip-lock. To do this, the unit was fully inflated with nitrogen to slightlyabove atmospheric pressure and then injected with the contents of a can ofspray used to dust off photographs. The gas contains Freon, and any leakagefrom the bubble would trigger beeping from a halogen leak detector. No leakagewas found when the detector was passed over structural seams, but the testrevealed that a portion of the zip-lock seal was not tightly closed and had to bereset. Testing with Freon was done only once to check the structural integrity ofthe bubble. It is not used as a seal check prior to each fumigation because thepresence of halocarbons interferes with the accurate operation of the trace-oxygen analyzer. To test for leaks before most treatments, a slight vacuum isdrawn on the enclosure at the beginning of each run and maintained for onehour. A system with large leaks would not be able to hold the vacuum. The sec-ond step in preparation for routine operation of the bubble was determining therate of oxygen permeation into the chamber through its walls. This was done inconventional fashion, except that the chamber was partially evacuated prior toadding nitrogen containing 10 ppm oxygen. The evacuation and nitrogenrecharging cycle were repeated until the oxygen concentration was reducedbelow 1000 ppm. The inflated bubble was allowed to stand for 24 hours whilethe oxygen concentration was followed on a Teledyne trace-oxygen analyzer.This established the rate of oxygen permeation, which can be expressed in ppmper day or per hour. Three runs revealed a rate of 800 ppm oxygen per day or33 ppm per hour (Fig. 7.4). This means that with a chamber half full of objects(which would cause the oxygen buildup rate to be 66 ppm per hour) and havingan atmosphere with 200 ppm oxygen, it would take about 12 hours to get to amaximum oxygen level of 1000 ppm. Depending on the volume of the loadbeing fumigated and the gastightness of the zip-lock seal, the time betweenpurging cycles could be as short as 8 hours or as long as 24 hours, but generally


Figure 7.4

Leak-rate measurement of oxygen concentra-tion vs. time in Houston's Museum of FineArts bubble.

it was found to have a range of 10-14 hours. Had the bubble been operated atoxygen levels as high as 3000 ppm, purge cycles might have been spaced asmuch as two days apart.

In a typical run, objects are loaded into the chamber with the steam humidifierand the fan, which are positioned to avoid water vapor directly over the materialbeing treated. The humidifier is filled with distilled water, and the oxygen ana-lyzer is calibrated at 20.9%using air. The humidifier and fan are turned on onlyduring the nitrogen entry period of the purge cycle. After the bubble is closed, aslight vacuum is drawn to remove some air (it is critical that the vacuum not beallowed to become excessive), and nitrogen is injected at no more than 60 psiuntil the bubble becomes full. The fan and humidifier are turned on again duringthis period. This completes one cycle. The nitrogen, fan, and humidifier are thenturned off, and evacuation starts the second cycle. Cycles are repeated until anoxygen level of 200 ppm is reached. The system is rechecked for gastightness.The oxygen level will slowly rise, and when it reaches 1000 ppm, the purgingroutine is repeated.

Steven Pine is a highly creative craftsman. His perfection of this anoxia systemand its frequent use at the Museum of Fine Arts in Houston demonstrate thatthe large Rentokil bubble can be used with nitrogen for the disinfestation ofmuseum objects. The process is labor intensive, requiring purging cycles every10-14 hours to keep the oxygen level below 1000 ppm. However, the bubblehas been used to treat as many as twenty-four upholstered chairs or over onehundred rolls of textiles at one time, and the effort is considered well justified.The availability of low-cost, high-purity gas in the Houston area keeps expensesfor nitrogen at a reasonable level.

Rigid Chambers

Homemade Containers

While a number of institutions carry out treatments in rigid and substantial plas-tic or metal containers, there are no units commercially available in the sense ofbeing listed specifically for nitrogen anoxia. However, there are many commer-cial chambers or containers for other uses that could readily be adapted foranoxia. Museum personnel, with considerable skill and ingenuity, have builtanoxia chambers out of pipes, boxes, water tanks, and units designed for usewith toxic pesticides. Chambers formed from a single molded piece of plasticand an easily attached cover have generally worked well, while boxes assembledfrom flat panels and joined by gasketed rims have not. Koestler (1992) even builta treatment system out of a heavy-gauge polyethylene utility storage cart. A

62Chapter 7

Oxygen ppmin nitrogen

Time, in hours

Plexiglas cover was sealed to a polyethylene chamber with silicone adhesive tocreate a container with an oxygen leak rate of only 50 ppm per day, which couldbe brought to an initial oxygen concentration of 50 ppm. Unfortunately, a newseal had to be prepared for each application.

Alan Johnston, of the Hampshire County Council Museum Service in the UnitedKingdom, put together a similar container (Fig. 7.5) using inexpensive off-the-shelf materials. It consisted primarily of a commercial polypropylene water tank,measuring 1 × 0.5 × 1 m, and an oval lid cut from a 10 mm thick sheet of castpolycarbonate. Johnston (1996) was able to solve the sealing problem with ordi-nary woodworking C-clamps, a large neoprene O-ring designed for aircraftdoors, and a smooth rim on the water tank. Nitrogen, humidified by passagethrough a three-bottle bubbler system, is fed into the chamber through a one-way truck tire valve set into the polycarbonate lid. This is connected to perfo-rated rubber tubing that coils around the base of the tank and distributes thenitrogen. The gas leaves the chamber through a simple valve system eitherdirectly to the atmosphere or through a Teledyne oxygen monitor. The valvearrangement allows either evacuation or purging. Although the system can bebrought to an oxygen concentration of 200 ppm when the unit is closed tightly,oxygen levels will climb to 1% over 168 hours. The system is most convenientlyoperated with the exit valves slightly open to allow a small flow of nitrogen tomove through the tank. RH and temperature readings are made using a Meacoradio telemetric sensor within the chamber that transmits data to a central envi-ronmental monitoring computer. RH is also followed in the third mixing jar ofthe bubbler system with a simple digital Meaco meter.

Steven Pine, of the Museum of Fine Arts in Houston, has also designed and con-structed several different types of rigid chambers for special applications. Heconverted a 20 ft (6.1 m) length of standard 15 in. (38 cm) diameter PVC pipeinto a treatment chamber for long and awkward objects like spears and rolls offabric (Fig. 7.6). Building this unit required a section of SDR 35 PVC gasketedsewer pipe, two 15 in. (38.1 cm) PVC caps, two 0.5 in. (1.27 cm) diameter


Figure 7.5

Diagram of custom-made anoxia chamber

used by Alan Johnston. (Courtesy of Alan

Johnston.)

Side elevation

Plan of chamber rim

End elevation

Figure 7.6

Anoxia chamber made of PVC sewer pipe.

(Photo courtesy of Steven Pine, The Museumof Fine Arts, Houston.)

silicone O-rings, and eight latches. First, the flange end of the pipe was cut off,then each PVC cap was cut perpendicular to the axis to provide both a collar anda half cap. Collars were mounted at each end of the pipe with PVC cement, andthe half caps were attached with latches and made gastight to the collars withthe O-ring gaskets. Humidified nitrogen is brought in at one end of the pipethrough a cap and out the other end through the second cap, either directly tothe outside or through an oxygen monitor (Fig. 7.7).

Figure 7.7

Diagram of improved PVC prototype. (Courtesy of Steven Pine, The Museum of Fine Arts, Houston.)

Materials(1) 15 in. × 20 ft (38 cm × 6.1 m) SDR35 PVC gasketed sewer pipe(2) 15 in. (38 cm) PVC gasketed sewer caps (two)(3) 0.5 in. (12 mm) silicone gasket O-rings (two)(4) Teledyne oxygen monitor

Chapter 7 64

Usual humidification system

Digital hygrometer placed in third jar allows forRH reading of the N2 entering the tube.

Large end of pipe is cut off.

A cap cut in half yields a half cap and a collar,which is welded onto the pipe with PVC cement.

Humidified N2

Input

Four latches on each cap

Half cap

O2 monitor

T-valve N2 outlet

O-ring gasket made of silicone

Any tightly sealable container with substantial walls that allow the installation ofgas and utility lines should be convertible to an effective anoxia chamber.Hardigg Industries makes such reusable containers in a range of sizes, and Pinehas converted a 9 ft3 (0.3 m3) case, shown in Figure 7.8 with gas ports and sen-sor connections, into a compact system with nitrogen supply and humidification(Fig. 7.9). This is easy to build and effective for the disinfestation of smallobjects in a standard dynamic process of allowing nitrogen to flow through thecontainer for 5-20 days as needed, after the oxygen content has been loweredbelow 1000 ppm.

Vacufume Chambers

Conversion of commercial metal fumigation chambers originally designed for usewith gaseous pesticides can provide a unit in which anoxia with high-puritynitrogen is effectively carried out. An example of this is the VacufumeFumigator-36, which was manufactured by the Vacudyne Company. Severalthousand have been sold, mostly to hospitals for sterilization treatments, but afew were acquired by cultural institutions like the Los Angeles County Museumof Art (LACMA) and the Royal Ontario Museum in Toronto. Conversion of thischamber for anoxia treatments consists primarily of installing a humidificationsystem and providing gastight entry to the box for humidity and oxygen moni-toring cables. Vacudyne offers conversion kits, and a number of museums havemade the change. LACMA purchased a Vacufume chamber in 1983 for use withethylene oxide at about the time that this fumigant fell out of favor. Conse-quently, the unit was never used with ethylene oxide. Researchers at the GettyConservation Institute proposed that the Vacufume chamber be converted tooperate by anoxia. Shin Maekawa and Vinod Daniel of the GCI, along withSteve Colton of LACMA, designed and made the necessary modifications. TheVacufume Fumigator-36 is a large metal box with an inner capacity of about1.02 m3. It measures 0.91 m square at the base and is 1.22 m high. The largegasketed door is sealed with four wheel latches, one at each corner, and whentightly closed, the chamber has an oxygen leak rate of about 200 ppm per day.The ability to keep oxygen inflow to this level is a major reason that the Vacu-fume unit is an attractive candidate for conversion.

Figure 7.8

Hardigg case with taps. (Photo courtesy of

Steven Pine, The Museum of Fine Arts,

Houston.)

Figure 7.9

Anoxia system based on Hardigg case. (Photo

courtesy of Steven Pine, The Museum of FineArts, Houston.)

Reusable Anoxia Systems 65

Infestations in the LACMA collection are not common, but the chamber findsfrequent use treating new acquisitions and incoming pieces on loan with nofumigation history. In a typical treatment carried out in April 1996, the chamberwas loaded with a large, well-wrapped, papier-mache construction held in acardboard box and a small painting in a wooden frame that showed signs ofinfestation. After the door was closed, the latches were secured tightly, and avacuum of 5 psig was pulled. The purpose of the vacuum is to remove someoxygen, to determine if the seal is holding, and to ensure that the chamber ismade as gastight as possible. The reading on the oxygen monitor fell to 17.5%.This showed that there was a good seal and that it was possible to turn andtighten the latches just a bit more. The humidification system, consisting of threestout, half-gallon, polypropylene bottles, was attached to the left outside wall ofthe chamber as shown in Figure 4.1. Humidification is provided and measured asdescribed in chapter 4. Nitrogen first sweeps out the air in the bubbler systemvia the purge line just before the entry valve to the chamber. Then the nitrogenis sent into the chamber, bringing it up to atmospheric pressure. A vent from thechamber is then opened to allow the unit to be flushed with nitrogen to achievethe required level of oxygen. In a typical run, the RH in the humidification sys-tem would quickly rise to 65%. This is higher than the RH in the chamberbecause the pressure there is higher. After the gas streams in the bubbler systemwere adjusted to send less nitrogen through the water bottle, the RH in the thirdbottle fell to 57%, and the RH in the chamber leveled at about 50%. The cham-ber is typically operated with the RH at 45-55%. Flushing the system over threeeight-hour shifts with 99.995% nitrogen brought the oxygen concentration to600-800 ppm. The chamber allows 200 ppm oxygen to leak in over twenty-fourhours, so sealing it at this level means that the oxygen level will drift up to under3000 ppm, or 0.3 %, over ten days. This is within the allowable increase in oxy-gen concentration for an effective anoxia treatment.

The Vacudyne Company has been testing a prototype of another kind of rigid-chamber anoxia system, offering the possibility that a turnkey unit will be madeavailable commercially. The system uses a conventional 50 gallon (190 liter) steeldrum for the anoxia chamber and comes with a three-jar humidification systemas well as the necessary gas and utility lines. The unit has been evaluated byMark Roosa, chief preservation officer for the Huntington Library Art Collectionsand Botanical Gardens in San Marino, California. The system will purge down toan oxygen concentration of 2000 ppm in 6-8 hours using nitrogen containing20 ppm oxygen. Treatment is generally run at 2000 ppm oxygen, 20 °C, and45-50% RH for up to 7 days, depending on the nature of the infestation. At theHuntington Library, treatment is done only where infestation is seen or highlysuspected. Pests have included silverfish, lice, beetles, and termites, which haveattacked books, masks, papier-mache objects, photographs, and other archivalmaterial. This is a usable and relatively inexpensive system that can be trans-ported to sites as needed. The main limitation may be one of capacity.

Chapter 7 66

Chapter 8

Fumigation withCarbon Dioxide

Although cultural institutions use anoxia more widely than they do carbon diox-ide fumigation to control infestations, the quantity of material treated by carbondioxide fumigation is larger. When big reusable chambers are employed to treatlarge batches of material, carbon dioxide is the gas of choice because gas com-positions effective for killing insects are more readily and cheaply achieved withit than with nitrogen. It is faster and easier to replace 60-80% of the air in a30 m3 bubble with carbon dioxide than it is to purge the chamber to an oxygencontent of 500 ppm with high-purity nitrogen and hold it at that level. Carbondioxide is preferred over nitrogen because it requires less stringent sealing condi-tions and because fluctuations in concentration have much less influence on per-formance. Carbon dioxide is not used with pouches and small containers, whereargon and nitrogen work well. Many institutions use both carbon dioxide fumi-gation in large reusable bubbles and nitrogen treatments in bags to deal withinfestations. Other reasons for favoring carbon dioxide for large-volume opera-tions are its much lower cost than nitrogen on an equal volume basis and thelesser need for auxiliary humidification. A nitrogen purge sweeps out all ambientmoisture from a treatment container, but at a 60% carbon dioxide level, 40% ofthe original water vapor remains. In addition, large chambers are often filledwith objects in wooden and cardboard boxes packed with paper, which providesa considerable amount of moisture buffering. There is some concern about theformation of carbonic acid when carbon dioxide encounters water during treat-ment as pointed out by Reichmuth (1987). The carbonic acid provides a pH of 4,which might damage sensitive materials such as metal filaments, zinc objects,and organic pigments. These could be harmed by the acidity, and conservatorsshould be aware of this factor when planning treatment. However, no problemsof this type were noted in the many reports on carbon dioxide fumigationreviewed for this chapter.

Effectiveness of Carbon Dioxide versus Nitrogen

To the extent that a direct experimental comparison can be made, nitrogencontaining very low concentrations of oxygen appears to be the more lethalmedium. In a study comparing nitrogen, argon, and carbon dioxide, Valentín(1993) observed that there was very little mortality among Hylotrupes bajulus,especially old larvae and pupae, held in an atmosphere of 60% carbon dioxide at30 °C for 3 weeks. Under the same conditions with nitrogen containing 300 ppmoxygen, total mortality of all life stages of the borer is achieved in 10 days.Valentín found that for most beetle pests found in museums, exposure to 60%carbon dioxide for 10-25 days at 30-35 °C was needed, while nitrogen contain-ing 300 ppm oxygen required only 4-6 days. She concluded that when compar-ing nitrogen or argon with carbon dioxide under typical fumigation conditions,the carbon dioxide atmosphere was ineffective for controlling populations ofCerambycidae (boring beetles), and elimination of species in the Anobiidae andLyctidae families took more time than with nitrogen. Banks and Annis (1990)attempted to make a comparison based on the limited available literature andconcluded that none of the gases has a marked advantage. However, the twostudies they cited (Jay 1984; Reichmuth 1987) both describe the use of nitrogencontaining 10,000 ppm (1%) oxygen, which is essentially ineffective for anoxia.Therefore, conclusions that nitrogen is no more effective than carbon dioxidewere erroneously based on studies made with high oxygen concentrations.

Carbon Dioxide Fumigation Requirements

In the current practice of conducting carbon dioxide fumigations in largebubbles, operators make up for lower effectiveness of this gas by running treat-ments under more severe conditions, that is, over longer times and at highertemperatures, but normally staying at 60% carbon dioxide. Treatments generallylast 2-5 weeks at temperatures above 25 °C. The use of carbon dioxide at 60%

67

Chapter 8

concentration derives from earlier studies involving this gas in treating storedgrains. Davis and Jay (1983) describe a number of successful field studies insilos, going back to 1973, where the concentration of the carbon dioxide wasgenerally 60%. Banks and Annis discuss recommended dose rates in some detail.Again, 60% is favored, but if more resistant species are present, such as ciga-rette or khapra beetles or rice weevils, up to 80% carbon dioxide may berequired to control all life stages within a reasonable exposure period. Smith andNewton (1991), after surveying the literature up to 1991, state that 60% is theconcentration generally considered to be most practical for control of storedproduct insects. Higher concentrations do not confer any great advantage andare more difficult to maintain. They confirmed this in their laboratory with stud-ies at 60%, 80%, and 100% concentrations. Recently published case studiesfind that fumigations typically are carried out at 60% carbon dioxide levels.Cipera and Segal (1996) normally use this dosage but increase the concentrationto 80% when dealing with wood-boring insects. Warren (1996), who conductsfumigations in a custom-built 50 m3 bubble, starts treatment with the carbondioxide concentration at 80%. This is allowed to drop to 60% over a 10-dayperiod and then is maintained at this level. The state-of-the-art unit nowmarketed by Rentokil automatically releases carbon dioxide until a 60% con-centration is reached.

Duration of exposure is highly dependent on temperature. In 1984 Jay recom-mended the following time-temperature profile for fumigation with 60% carbondioxide: 21-28 days at 16 °C; 10-14 days at 21 °C; and 4-7 days at 27 °C. In1987 Annis claimed that the time required for the control of most grain-infestingspecies at 60% carbon dioxide and 20-29 °C is about 11 days. Annis chose thiswide temperature range so that he could review data from a collection of trialscarried out under diverse conditions. However, Jeremy Jacobs (1996), workingwith natural history collections at the Smithsonian Institution, found that some-what more control of temperature was required for effective treatment of thespecies that were plaguing his collection, particularly odd beetles (Thylodriascontractus). Jacobs was using carbon dioxide for fumigation in a large Rentokilbubble. His initial program was designed to operate with 60% carbon dioxidefor 15 days at 30 °C. Fifteen runs were made over a two-year period. The carbondioxide level fluctuated above and below 60%, but never lower than 55%, whilethe temperature varied in a range lower than the desired 30 °C, generallybetween 19 and 27 °C. When either parameter fell too low, Jacobs attempted tocompensate by running the treatment for 15 more days or longer. Success wasdetermined by returning items to their storage cases, which had been given apyrethrum-silica gel treatment while the objects were in the bubble, and moni-toring very closely for any sign of reappearance of insects over the ensuingtwenty-four months. A reemergence of insects occurred in 40% of the runs.Jacobs examined his operating parameters against these results and noted thatgetting 100% mortality was not related to carbon dioxide level nor to length oftreatment. However, a complete kill was obtained in every run where thetemperature was 25 °C or higher. Getting to a suitable temperature can be aproblem for installations housed in poorly heated quarters during cold seasons.Both Jacobs and John Burke, who operates a similar system at the OaklandMuseum of California, do not carry out treatments in the winter.

Smith and Newton (1991) described studies designed to provide minimumassured kill times for carbon dioxide fumigation with the same goal that Rustand Kennedy had for nitrogen anoxia. Table 8.1 provides the minimum numberof days of exposure to a 60% carbon dioxide atmosphere required to obtain100% mortality of cultures containing all life stages of fourteen species ofinsects or mites. Experiments were carried out for 2 and 4 weeks at 15 °C, whileat 23 °C and 35 °C they were generally run for 1, 2,4, 8, and 14 days in order to

68

Table 8.1Minimum time for assured 100% kill of all

stages of property-destroying insects and mitesby use of 60% carbon dioxide at 75% RH.

find the minimum time needed for complete kill. Cultures containing insect eggs,pupae, adults, and young and mature larvae were obtained by placing samplesin 100 g of standard rearing medium in half-liter jars and incubating under stan-dard conditions. Mite cultures were inoculated during the week before treat-ment with 5 ml of live culture medium. The jars were stacked in a high-densitypolyethylene container with a 35 l capacity. After closing, it was purged withcommercial carbon dioxide until the concentration reached 60%. Carbon dioxidewas added as needed to maintain this level. The RH was kept at 75% to promotethe survival of mites and book lice. One culture of each species was exposed toeach combination of temperature and treatment duration. Each culture was

69Fumigation with Carbon Dioxide

Mold mite

Grain mite

American cockroach

Khapra beetle

Rice weevil

Varied carpet beetle

Merchant grain beetle

Red flour beetle

Cigarette beetle

Hide beetle

Australian spider beetle

Tropical warehouse moth

Book louse

Furniture beetle larvae

2335

15

23

35

15

23

35

35

15

25

35

15

25

35

15

23

35

15

23

35

15

23

35

15

23

15

23

35

15

23

35

23

15

23

35

24

14

14

14

1

14

2

4

14

28

14

4

28

14

2

14

4

1

14

4

4

14

4

1

14

4

14

1

4

14

4

2

8

14

1

14

Species Temperature (°C) Time (days)

examined two days after treatment and again after a holding period longenough to permit the development of visible life stages. Complete mortality wasdeclared only if there were no signs of life at both inspections. Mite and booklouse cultures were examined with a low-power microscope.

The teams of Smith and Newton and Rust and Kennedy determined minimumtimes for assured complete kill with two different fumigants. Both studiesshowed the same relative susceptibility of species to dehydration, but theanomalies in the work by Smith and Newton indicate that it was not done withthe same rigor as the research by Rust and Kennedy.

Rentokil, Inc., is an international corporation in the United Kingdom that dealswith many aspects of pest management. Through its Project Development Unitin West Sussex, under the direction of general manager Colin Smith, the com-pany has in recent years focused on designing and promoting large reusableplastic chambers, called bubbles, for insect disinfestation. These were initiallyused with conventional fumigants like methyl bromide but more recently withcontrolled atmospheres, particularly carbon dioxide. John Newton (1990) hasbeen heavily involved in Rentokil's research programs. The Rentokil bubbles aremanufactured by Power-Plastics Limited in Thirsk, North Yorkshire, England, acompany best known for its production of tarpaulins and tents. Fully equippedand automated units are sold and serviced in North America by Maheu &Maheu, Inc., of Quebec, Canada, and elsewhere by Power-Plastics Limited.

Carbon dioxide fumigation has a long history of use for killing insects infestingstored grains. One of the earliest demonstrations of its use for treating museumitems was work with costume collections in the Museum of St. John in London(Child 1988). Textile pests were successfully eradicated with 60% carbon dioxidein a standard Rentokil bubble. Since then, carbon dioxide fumigation hasbecome popular among museums in the northeastern United States and inCanada. There are several reasons for this. Institutions in this geographical areahave a need for large-volume fumigations; Canada has favorable regulations onthe use of carbon dioxide for treating insect infestations; and Maheu & Maheu,supplier of the Rentokil bubble, backs up its sales efforts with strong technicalsupport. The Canadian government does not classify carbon dioxide as a fumi-gant but as an inert gas. Therefore, no special license is required to use it. Incontrast, the British government now lists carbon dioxide as a pesticide. As aconsequence, this gas is not likely to find much use by conservators in theUnited Kingdom. In the United States, there appears to be no federal policy onthe handling of carbon dioxide, and each institution is required to check with stateregulatory authorities.

Treatment Experience in the United States

In the late 1980s, when the most important toxic fumigants—such as ethyleneoxide—were being banned in many areas, several institutions in the north-eastern United States began looking at carbon dioxide as a safe fumigation sub-stitute that could disinfest large quantities of material. The Winterthur Museumin Delaware, the Society for the Preservation of New England Antiquities(SPNEA), and Old Sturbridge Village in Massachusetts each have very largeholdings of decorative arts, historic furniture, and objects of paper, wood, andtextile that are prone to attack and infestation, generally by powderpost and car-pet beetles, silverfish, and clothes moths. At that time, SPNEA had forty-fourhouses filled with rooms of furniture under its control, while Old SturbridgeVillage owned forty-two historic buildings holding thirty-eight thousand booksand eighty thousand objects.

70Chapter 8

Old Sturbridge Village

Old Sturbridge Village was the first of these institutions to act. The nature andsize of its holdings create a substantial need for dealing with insect infestations.Webbing clothes moths and carpet beetles are the most common problems. Incontrast to almost everyone else who is using carbon dioxide fumigation in aplastic bubble, Old Sturbridge Village carries out its treatments in a rigid metalchamber originally purchased in the early 1980s for operation with ethyleneoxide. In 1989, with a few modest changes in the external plumbing, it wasconverted to use with carbon dioxide. The chamber, with a 2 m3 capacity, canhandle relatively small loads, but the conservators at Old Sturbridge Village havebeen able to turn to their colleagues at SPNEA for help with larger objects, suchas sofas. The rigid chamber makes it possible to draw a slight vacuum, whichspeeds the purging process when a new run is started. The chamber is filled withcarbon dioxide by two displacements. Instead of humidifying the gas stream,objects are conditioned to 45% RH before being put into the chamber. Aninfrared safety monitor outside the unit trips an alarm if the outside carbondioxide concentration reaches 1000 ppm.

An efficient heating-evaporation device, placed between the gas cylinder andthe chamber, is an important part of any carbon dioxide fumigation system. Thecarbon dioxide coming from the cylinder must evaporate and expand rapidly; inso doing, its temperature drops sharply. To avoid ice formation and other prob-lems and to bring the carbon dioxide stream into the chamber at close to thedesired run temperature, heat must be supplied at the point where the evapora-tion occurs. To do this, the Sturbridge system uses an in-line electric heater,E99-E-Heater 320, which is supplied by the Air Products and ChemicalsCompany. It is thermostatically controlled to operate between 24 and 29 °C(75 and 85 °F).

The Winterthur Museum

In 1988 Rentokil representatives brought a small plastic chamber to theWinterthur Museum and successfully demonstrated the ability of carbon dioxidefumigation to rid trial objects of living insects. As a result, a standard 30 m3

Rentokil bubble was purchased and installed in 1990 in an isolated area of themuseum that had previously been dedicated to the use of toxic fumigants.Winterthur adopted a procedure that requires 14 days of treatment with60-70% carbon dioxide at ambient temperature,, which is normally 20-25 °C.The carbon dioxide is not humidified. Although this has been a concern, result-ing in close inspections of treated objects, no damage due to not adding watervapor has been found. Rugs and upholstered furniture have been the mostroutinely treated items. Fumigations have been done primarily as a precautionrather than as a response to observed infestations. Objects, as they are receivedor sent out, are routinely examined for the possibility of infestation and the needfor processing. After several years of operation, the bubble has been needed lessfrequently and treatment schedules have been erratic. The unit was not used atall in 1995.

Society for the Preservation of New England Antiquities

In 1991 SPNEA had a burgeoning infestation problem. One of the historichouses under its management had become infested with webbing clothes moths.In addition, the house accidentally received an internal covering of soot from amalfunctioning furnace. Cleaning away the soot, exposure to a cold NewEngland winter, and some contact spraying with insecticide eliminated about95% of the infestation. However, rolls of textiles in the society's storage facilitywere also infested, and a standard Rentokil bubble was purchased and


immediately put into service in 1992. The stored textiles and other remainingproblems were completely rid of their insect pests in the chamber, using carbondioxide at 20 °C. Typically, the carbon dioxide level is maintained between65% and 75%. This is monitored by measuring the oxygen level, which is keptbetween 4.9% and 8.4%, a spread that corresponds to a carbon dioxide rangeof 75-60%. Humidity is provided by a bubbler system that takes the unit to40% RH in the summer but only 30% in the winter.

Only two weeks of treatment are required to eradicate powderpost beetles andclothes moths in textiles, even when these are wrapped in plastic film. However,on three occasions adult carpet beetles survived two weeks of fumigation, andthe standard run time is now three weeks when this species is found. Veryheavily infested objects may be subjected to two treatments. Longer run timesand repeated treatments might have been avoided by operating the system at25-30 °C. About four to six T-cylinders of gas are used per run. The carbondioxide goes through a system of two gauges and a heat exchanger, which wasobtained from the gas supplier. The carbon dioxide concentration in the roomoutside the bubble is monitored by an infrared analyzer, which is set to triggeran alarm and turn on a powerful ventilation system when safe levels areexceeded. The ventilation system is also used to exhaust carbon dioxide fromthe bubble and draw in air at the end of a run. Over a four-year period leadinginto 1996, thirty-five treatments, averaging about one hundred objects each,had been put through the three-week process without any apparent failure.SPNEA also does work for other institutions. Perhaps the most unusual assign-ment was the treatment of bales of hay used to make a sculpture for display inthe Portland Museum, in Maine.

The Oakland Museum

A grant from the Institute of Museum Services enabled John Burke to purchase astandard Rentokil bubble in 1991 and install it in the conservation facility of theOakland Museum, in California (Fig. 8.1). The unit obtained by Burke has a ceil-ing adjustable to a height of 2.44 m (8 ft). One side contains a plastic window,and a gas-transfer manifold is in place along the bottom of an adjacent side.Burke had intended to use nitrogen anoxia as his method of treatment. After

Figure 8.1

Room-size carbon dioxide bubble. (Photocourtesy of John Burke, Oakland Museum.)

Chapter 8 72

several months of manipulating the bubble, however, he concluded that a nitro-gen atmosphere with less than 1% oxygen could not be maintained in the sys-tem because of high leakage rates. Accordingly, the system was altered for usewith carbon dioxide. Treatment is carried out by displacing enough air with purecarbon dioxide to establish and hold the concentration of this gas at 65-70% forrelatively long periods of time. The system was first used to disinfest the entireethnographic collection of the museum over six months using 4-week exposuresessions. Subsequently, large batches were treated for 5 weeks for both in-houseprojects and for other clients. Because of his work, Burke introduced the practiceof both nitrogen anoxia and carbon dioxide fumigation to a large number ofconservators in the San Francisco Bay Area very early in the development of thistechnology. Large batches of material are now cleared of invading insects in hischamber five or six times a year. These include ethnographic collections, biblio-graphic materials, textiles, sculpture, and furniture and other large pieces thatare placed in the chamber in a loose arrangement. Smaller objects are packed inboxes, and collections are stored in drawers and cabinets (Figs. 8.2 and 8.3).When objects are treated, the materials are placed in the chamber, and the ceil-ing is lowered to eliminate as much free volume as possible. First, a slightvacuum is pulled, and then carbon dioxide is flushed into the chamber through aheat exchanger. The cycle is repeated several times until a carbon dioxide levelnear 70% has been achieved. The concentration is determined with an oxygensensor connected to a digital readout, which is calibrated to show increasing car-bon dioxide levels as the oxygen level falls. Because there is leakage in the sys-tem, the carbon dioxide level is logged on a daily basis and brought up to 70%as needed. Sensors placed inside closed cardboard boxes have shown that car-bon dioxide concentrations inside these containers usually rise to free-spacelevels in less than two hours. No attempt is made to add water or otherwisecontrol humidity.

On two occasions initial treatment did not provide 100% kill, and this wasremedied with another run through the process. Burke has found that warm, dryconditions, normal at his location, are most effective for achieving a rapid kill,while cold, damp weather inhibits the process. Thus, he tends to avoid makingruns during the winter months. The five-week treatment period might seemexcessive, but the success achieved with this process over four years has createda body of data showing that 35 days of 65-70% carbon dioxide exposureis effective.

The Smithsonian Institution

The Smithsonian Institution began treating parts of its holdings with carbondioxide in 1993. Other than the fact that emphasis has been on natural historycollections rather than on fine arts or decorative arts, the issues behind its acqui-sition of a fumigation bubble are not unlike those faced by other institutions.The Smithsonian has over 580,000 mammal specimens, the largest collection inthe world, housed in several large and separate rooms. Odd beetles andredlegged ham beetles were known to have infested the collection, but thesewere believed to have been held in check by fumigation based on carbon tetra-chloride done in the 1980s. No integrated pest management was employed, andit was not known how effective the fumigation program was. The use of toxicchlorocarbons in the museum was finally halted in November 1990, and anintegrated pest management program was started. In 1992 the Smithsonianpurchased a Rentokil bubble, measuring 2.7 m square and 2.4 m tall, that wascustom-made to fit into available building space. Following a period of installa-tion and trial, treatments began in September 1993. After fifteen runs, madeover a two-year period by museum specialist Jeremy Jacobs, an optimum

73

Figure 8.3

Collections in steel cabinets. (Photo courtesyof John Burke, Oakland Museum.)

Fumigation with Carbon Dioxide

Figure 8.2

Collections in open trays. (Photo courtesy of

John Burke, Oakland Museum.)

procedure was developed requiring fumigation with 60% carbon dioxide at 30 °Cfor 2 weeks.

Treatment Experience in Canada

The National Museum of Science and Technology

The National Museum of Science and Technology in Ottawa has an outstandingcollection of historic vehicles comprising railcars, automobiles, and horse-drawncarriages and sleighs, all very large museum objects. Some are on display, butmost are kept in warehouse storage. The fabrics of these vehicles are prone toinfestation by casemaking clothes moths and, occasionally, by carpet beetles.The presence of moths in the collection had been noted in the early 1980s, whena number of inadequate control steps were initiated. By 1989 it was clear to SueWarren, conservator at the museum, that the infestation had spread through thewarehouse. She turned to Thomas Strang of the Canadian Conservation Institutefor help. He was an important early advocate of carbon dioxide fumigation andpromoted its use particularly with museums in Canada. Strang provided supportand guidance to conservators and responded effectively to concerns about theuse of carbon dioxide, such as the formation of carbonic acid.

About the time that Warren contacted Strang, he was looking for a location totest his ideas and recommended carbon dioxide fumigation to the NationalMuseum. A large, custom-built bubble would be needed to handle the museum'ssizable vehicles, but before the museum made this commitment, it leased a stan-dard 30 m3 Rentokil chamber from Maheu & Maheu to evaluate the system andits effectiveness. Two fumigations were carried out according to the proceduresdescribed below. Four small, moth-infested sleighs and control groups of tenlive adult moths and ten larvae were placed in the bubble for each run. A com-plete kill was obtained for both control groups, and the sleighs appear totallyfree of infestation after five years. With proof in hand, the museum purchased acustom-built bubble measuring 6.6 × 2.44 × 2.74 m. This bubble will hold themuseum's largest horse-drawn carriage, two medium-size vehicles, or foursmall sleighs.

In addition to the bubble, the system contains a vacuum pump, an ultrasonichumidifier, a small fan, a data logger for recording temperature and humidity,and a carbon dioxide monitor with remote sensor. Warren decided to include ahumidifier when she noted that unless moisture were added, small cracks orsplits in old wooden paneling would widen over a 2-week fumigation, althoughthis was not observed to make much of a difference on relatively large splits.Movement around loose paint or joints where there were fills was also very dam-aging. The museum made several additions to the chamber. Wiring is broughtinto the bubble through an acrylonitrile-butadiene-styrene (ABS) plumbing cap,which is sealed with silicone caulking. Small metal pulleys replace the plasticloops used to raise the top of the bubble from the base. A sturdy tarpaulincovers the bubble's ground sheet to minimize wear on the bubble floor. A rampover the zipper is used to facilitate rolling vehicles into the unit and to protectthe seal. With constant use over five years, the bubble, including the zipper, hasheld up well. The seal is maintained in good working condition by keeping thezipper clean and treating it with a silicone spray every three or four cycles. Twotears, both caused by nails protruding from vehicles, were easily patched.

For a treatment, the chamber is first flushed to a level of 80% carbon dioxide insix to seven hours. This is done by initially removing a portion of the air with thevacuum pump and then refilling the bubble by flowing carbon dioxide into itthrough two entry ports at a rate of 2 m3 (60 ft3) per hour. The concentrationdecreases by 2% per day but is not allowed to fall below 60%. Temperatures

74Chapter 8

range from 15 to 25 °C in spring and 20 to 24 °C in summer. The temperature inthe warehouse in winter is as low as 12 °C, and as is the practice of a number ofinstitutions, fumigations with carbon dioxide are not done during the cold sea-son. A typical cycle lasts 10-14 days. These conditions seem to be relativelymild, but clothing moths are quite susceptible to modified atmospheres.

After five years the list of treated objects includes ninety-two carriages andsleighs, five automobiles, one piano, and a large assortment of smaller objects.Following fumigation, the vehicles are thoroughly vacuumed and enclosed in asealed polyethylene bag to prevent reinfestation. Insect sticky traps are placedwith objects that had been more heavily infested. All stored items are examinedin the spring and in the fall when moths are most likely to be active. After all ofthis, only three findings of dead moths suggest that reinfestation may havetaken place. The conservators at the National Museum of Science and Technol-ogy have eliminated clothes moths from a large portion of their collection andfeel that their carbon dioxide fumigation program is successful.

The Canadian Museum of Civilization

The Canadian Museum of Civilization in Hull, Quebec, has an ongoing need tofumigate large quantities of material, including traveling exhibits and new acqui-sitions as well as parts of its permanent collection suspected of infestation. In1991, soon after contracting for US$40,000 to eradicate a major infestation ofclothing moths by freezing, the museum switched to carbon dioxide fumigationfor pest control. It purchased a standard 30 m3 Rentokil bubble from Maheu &Maheu and has found the procedure to be much cheaper than freezing. In thefirst two years, the bubble was used forty times at a total cost of US$12,000.This sum includes US$8,000 for the bubble, US$2,500 for an Armstrong (AM6-1011/1012) two-channel carbon dioxide monitor, and US$35 per run for car-bon dioxide. In addition to the economic advantage of carbon dioxide fumiga-tion, the conservators responsible for the work, Luci Cipera and Martha Segal,feel that the in-house operation provides better control of the process. There isless handling and movement of material and, therefore, less chance of damageto the objects. Another reason for choosing this approach is the versatility in thetypes, quantity, and size of artifacts that could be treated. Large objects thathave been easily accommodated in the bubble include 4.27 m (14 ft) woodenhouse posts in crates, entire displays of mannequins and props, and tree stumpsand large branches used in large scenic representations.

All of the museum conservators are trained to use the system, and personnel arescheduled on a rotating basis. The most effective and safe operation is achievedby using two people during each run. Depending on the frequency of fumiga-tion, which is limited to twice a month, up to six to eight days of personnel timeare required per month. This includes time spent on getting artifacts from stor-age, loading them into the bubble, filling the bubble with carbon dioxide, deflat-ing it after the run, and unloading and returning the artifacts. Typically, articlesare placed directly on the floor of the bubble or, for smaller pieces, on shelves.After the top is lowered and the plastic zipper is sealed, the bubble is inflatedwith air and left overnight as a leak check. The air is then pumped out andreplaced with an atmosphere of 60% carbon dioxide. Normally, the bubble iskept at that concentration and at room temperature for 2 weeks. As would beexpected, wood-boring insects have been found harder to exterminate, andobjects containing them are treated with 80% carbon dioxide for 3 weeks. Arecording hygrothermograph is used to monitor RH and temperature, but theseparameters change little during a run. Auxiliary humidification is not used, andthe RH is generally unchanged by the partial replacement of air with carbondioxide. Occasionally, the introduction of the dry carbon dioxide will cause theRH to fall at the beginning of a run. In a short time, however, humidity buffering


by the contents of the bubble brings the RH back to a level close to the initialvalue. Safety is a major concern. The bubble is normally placed in an isolatedroom originally designed to house a chamber employing toxic fumigants. Theroom has negative pressure, a ventilation system separate from the other sys-tems in the building, and a strong fresh-air exchange. The bubble is portable andcan be used elsewhere, preferably in an isolated, outdoor location where no oneelse is working. To follow carbon dioxide levels at critical points, an Armstrongtwo-channel carbon dioxide monitor is used. One sensor monitors carbon diox-ide inside the bubble, and the other is calibrated to measure carbon dioxide inparts per million in the room. This will set off an alarm when the level of carbondioxide there exceeds 1000 ppm. Other safety features include warning signson each entrance to the room and the limiting of access to personnel operatingthe bubble.

Carbon dioxide fumigation at the Canadian Museum of Civilization has beenhighly successful. In May 1996, museum conservators reported that there hadbeen no reinfestations occurring in the collections treated with this gas. Themuseum's success with carbon dioxide fumigation has also been helped with thestrong support and advice from sister institutions in Canada—the CanadianConservation Institute and the National Museum of Science and Technology. Theawareness of the seriousness of insect infestation has increased at the CanadianMuseum of Civilization, and this has led to a greater involvement on the part ofother divisions in a total, integrated program of pest management throughoutthe institution.

Treatment Experience in Germany

Perhaps the most extensive and creative application of carbon dioxide for pestcontrol has been done in Germany by professional exterminators like GerhardBinker. Binker is experienced in the use of the conventional toxic fumigants buthas also embraced nitrogen and carbon dioxide for a variety of applications. Hehas designed a series of procedures to meet needs on both small and largescales, from treating movable objects to fumigating sections of buildings wherecontainment involves using portions of the structure itself as container walls. Hehas even led the way in developing a technology for treating entire buildingswith carbon dioxide (Binker 1993a). Binker has developed an expandable gas-tight chamber, sold commercially as the "Altarion," to deal with movable loadsand objects of varying size. The capacity of this unit can be increased by addingchamber elements as needed. Binker claims that the system is able to control andrecord temperature, humidity, and gas composition for either nitrogen or carbondioxide treatment of museum objects and archival materials. The system can beoperated in locations where other activities are taking place if external gas sen-sors are carefully placed and monitored.

Binker treats infested objects that are very large or cannot be moved—forexample, high altars, pulpits, and church pews—by first covering them withimpermeable sheeting that is sealed against floor or wall sections and made asgastight as possible. Fumigation with carbon dioxide is then done in a conven-tional way. Binker found wood-boring beetles to be the major culprits attackingthe churches and historic buildings he has treated. Before doing a partial-building fumigation for these pests, it is vital to determine that the rest of thestructure is not infested. This is done with a complete and careful examinationusing established pest management procedures before fumigation is undertaken.It makes no sense to kill off these insects in one part of a building if adjacentareas are also infested.

An excellent example of carbon dioxide fumigation of an entire large building isthe Catholic church in Schaeftlarn, Germany, that Binker treated during the

76Chapter 8

Figure 8.4

Tonnage delivery of carbon dioxide. (Photo

courtesy of Gerhard Binker.)

Figure 8.5

Salmdorf Pietà undergoing treatment. (Photo

courtesy of Gerhard Binker.)

summer of 1995. The volume of space that had to be filled with gas, originally2.31 x 105 m3, was decreased with air-inflated balloons. This reduced theamount of fumigant that was needed. Still, an enormous quantity of carbondioxide had to be maintained at a concentration of 70% over the six weeks oftreatment. To meet this need, carbon dioxide in liquid form was periodicallydelivered by tanker and pumped into a large supply tank. It was automaticallywithdrawn from the tank and passed through a battery of heat exchangers toconvert it to room-temperature gas. A monitoring and control device measuredthe concentration of carbon dioxide in the church and automatically made up forlosses incurred by leakage during the extended treatment period. An unexpectedsource of leakage was through the floor and out the basement. Overall, a thou-sand tons of carbon dioxide were needed to complete the project (Binker 1996).Figure 8.4 is a photograph showing the magnitude of the carbon dioxide deliv-ery operation. The tanker on the right is unloading carbon dioxide into the mainstorage tank while a second tanker behind it waits its turn. Two banks of heatexchangers are visible on the left. Throughout the treatment period, the RHranged from 58% to 70% and the temperature from 19 to 29 °C.

A second example of Binker's work on a large scale is the fumigation of theCatholic church in Salmdorf, near Munich, where an important wooden religioussculpture, the Salmdorf Pietà, is housed (Fig. 8.5). The church was infested withwood-boring beetles, particularly furniture beetles, which had gotten into thealtars, pews, pulpit, and organ, as well as into the religious sculpture. After thebuilding was covered with large sheets of poly(vinyl chloride) film, sealed withtape, and made as gastight as possible, an additional stormproof tent was con-structed over the church, including the steeple (Fig. 8.6). This was done to mini-mize gas loss and to help retain humidity required for the protection of thewooden artifacts. A temperature and moisture sensor was placed on the Pietà toensure that dehydration was not occurring. There were heavy rains during thesealing of the church, boosting the RH. Still, additional humidification wasrequired to maintain a moisture level that the operators felt would protect thestatues from becoming too dry and cracking. Over the treatment period, the RHranged from 58% to 70% and the temperature from 19 to 29 °C.

In addition to worries about dehydration, there was concern that the carbondioxide might change the polychrome of the Pietà. A pigment panel was placednear the statue during fumigation and later compared to a reference panel. Nei-ther color change nor cracking of the statue were detected. The effectiveness of

Fumigation with Carbon Dioxide 77

Figure 8.6

Church tented for carbon dioxide fumigation.(Photo courtesy of Gerhard Binker.)

the treatment was determined by sealing larvae of the furniture beetle and theold house borer inside blocks of wood and placing them in the church duringfumigation. Subsequent examination found that all larvae were killed. Catholicchurches in Kranzberg, Unterschwillach, Peiting, and Gstadt/Chiemsee were alsosuccessfully treated with this approach.

Binker has also demonstrated a very rapid procedure that uses carbon dioxideunder pressure for dispatching insects. Objects containing infestations are placedin a pressure chamber and brought to a pressure of 300-600 psig. After two tothree hours, the system is depressurized, and 100% mortality is realized. Thisprocedure kills by a mechanism other than dehydration. The cells of insectsreadily absorb carbon dioxide under pressure; with rapid depressurization, thecell walls burst irreversibly and the insects die. Inert cellulosic material is notdamaged by this treatment since it cannot absorb large amounts of carbon diox-ide. Binker speculates that this procedure may, in the same manner, also kill fun-gal growth (Binker 1993b).

Chapter 8 78

Appendix A

ProfessionalContacts

Conservators

The following conservators and conservation scientists have substantial experi-ence using methods described in this book for the practice of insect control inmuseums. In many cases, they have also done important research and conductedworkshops in this field. All have expressed a willingness—indeed, even an eager-ness—to share their experience with other conservators and have dared torespond to requests for help.

John Burke, Chief of ConservationOakland Museum1000 Oak StreetOakland, CA 94607Tel: (510) 238-3806Fax:(510) 238-6123

Steve Colton, Objects Conservator755 Luton DriveGlendale, CA 91206Tel: (818) 247-2398Fax:(818) 548-6286

Vinod Daniel, Scientific OfficerMaterials Conservation DivisionAustralian Museum6 College StreetSydney, NSW 2000AustraliaTel: (61) 29320 6115Fax: (61) 2 9320 6070E-mail: [email protected]

Mark Gilberg, Research CoordinatorNational Center for PreservationTechnology and TrainingNSU Box 5682Natchitoches, LA 71497Tel: (318) 357-6464Fax: (318) 357-6421

Gordon Hanlon, AssociateConservatorDecorative Arts and SculptureConservationThe J. Paul Getty MuseumSuite 10001200 Getty Center DriveLos Angeles, CA 90049-1684Tel: (310) 440-7178Fax: (310) 440-7745E-mail: [email protected]

Jeremy Jacobs, Museum SpecialistDepartment of Vertebrate ZoologySmithsonian Institution10th and Constitution Avenue, N.W.Washington, DC 20560Tel: (202) 786-2550Fax: (202) 786-2979E-mail: [email protected]

Alan Johnston, Principal MuseumsOfficerProject Management and CollectionsCareHampshire County Council MuseumsServiceChilcomb HouseChilcomb LaneBar EndWinchester SO23 8RDUnited KingdomTel: (44) 962 846304Fax: (44) 962 869836

Dale Paul Kronkright, SeniorConservatorMuseum of New MexicoP.O. Box 2087Santa Fe, NM 87504-2087Tel: (505) 827-6358, ext. 634Fax: (505) 827-6349E-mail: [email protected]

Shin Maekawa, Senior ScientistThe Getty Conservation Institute1200 Getty Center DriveLos Angeles, CA 90049-1684Tel: (310) 440-6813Fax: (310) 440-7711E-mail: [email protected]

Steven Pine, Decorative ArtsConservatorMuseum of Fine Arts, HoustonP.O. Box 6826Houston, TX 77265Tel: (713) 639-7730Fax: (713) 639-7740

79

Appendix A 80

Gary Rattigan, PreparatorSociety for the Preservation ofNew England Antiquities141 Cambridge StreetBoston, MA 02114Tel: (508) 521-4788Fax: (508) 541-4789

Dr. Thomas StrangCanadian Conservation Institute1030 Innes RoadOttawa, Ontario K1A OMBCanadaTel: (819) 776-8417Fax: (819) 776-8300

Dr. Nieves Valentín, DirectorMinisterio de CulturaInstitute del Patrimonio HistóricoEspañolGreco 428040 MadridSpainTel: (34) 1 544 7894Fax: (34) 1 544 7778

Sue Warren, ConservatorNational Museum of Science andTechnologyP.O. Box 8724, Station TOttawa, Ontario K1G-5A3CanadaTel: (613) 991-3061E-mail: [email protected]

Professional Fumigators

The following individuals are professional fumigators experienced in the use ofnontoxic gases for insect control. They have worked with museum collectionsand are sensitive to the frailties and needs of fine art.

Werner BiebelBiebl und Söhne, Hygiene GmbHBergestrasse 882024 TaufkürchenGermanyTel: (49) 89 612 00 00Fax: (49) 89 612 00 037

Gerhard BinkerBinker Materialschutz GmbHPostfach 4D90571 SchwaigGermanyTel: (49) 911 5075011Fax: (49) 911 5075782

Michel MaheuMaheu & Maheu, Inc.710 Bouvier Street, Suite 195Quebec, QC G2J 1C2CanadaTel: (418) 623 8000Fax: (418) 623 5584E-mail: [email protected]

John Newton, Research EntomologistRentokil Group PLCResearch and Development DivisionFelcourt, East GrinsteadWest Sussex RH192JYUnited KingdomTel: (44) 1342 833022Fax: (44) 1342 326229

Colin Peter Smith, General ManagerProduct DevelopmentRentokil LimitedFelcourt, East GrinsteadWest Sussex, RH192JYUnited KingdomTel: (44) 1342 833022Fax: (44) 1342 326229E-mail: [email protected]

Appendix B

Materials andSuppliers

The following list of materials used to make the anoxia systems described hereinis not exhaustive. Other sources for most of the items include general hardwarestores, as well as laboratory and conservation supply houses.

Anoxia Systems (Custom-Made)Alan Johnston System

Suppliers

TankSolent Plastics136 Shirley RoadSouthhamptonUnited Kingdom

Tubes/Valves/FittingsAir and Hydraulic Hose ServicesUnit 29E Parnham DriveEastleigh, HantsUnited Kingdom

Oxygen AnalyzerTeledyne Brown EngineeringHarlequin CentreSouthall LaneSouthall, LondonUnited Kingdom

Perspex SheetAlda Plastics21A High StreetLyndhurst, HantsUnited Kingdom

Neoprene Seal O-RingBlue Diamond Bearings1 Bishopstoke RoadEastleigh, HantsUnited Kingdom

Nitrogen and RegulatorsBritish Oxygen Company

Anoxia Systems (Custom-Made)Steven Pine System

PartsSwagelok brass fittings: Union TeeB-600-3, Port Connector B-601-PC,Union Elbow B-600-9, Bulkhead UnionB-600-61, Ferrule Set B-600-Set,Nupro Valve B-6JN

Dow Corning silicone caulk, 732,as gasket material at the couplingsthat penetrate the containment walls.Neoprene gaskets and Teflon tape canalso be used.

81

Nalgone polyethylene bottles(half-gallon, Fisher 11-815-11A)

Gasketed PVC sewer pipe, 15 in.(38.1 cm) diameter by 20 ft (6.1 m)long

Gasketed PVC sewer pipe caps, 15 in.(38.1 cm) diameter, ½ in. (1.27 cm) thick

Silicone gaskets, 0.5 in. O-rings—round cross section for the Hardiggcase and square cross section for thepipe. From Amerflex Rubber and GasketCo. and many other sources.

Teledyne oxygen monitor cap—TBE/A1part no. 1406A51

Retroseal PVC and ABS cleaner and PVCsolvent cement

Cole-Parmer E-37500-06 panel-mounteddigital thermohygrometer

Sears 16 in. oscillating fan

Sears Duracraft Delux, 2.5 gal humidifier

Anoxia Tent Supplies(for Custom-Made Systems)

Supplier

McMaster-Carr Supply Co.9630 Norwalk Blvd.Santa Fe Springs, CA 90670Tel: (310) 692-5911

Most materials can be obtained fromMcMaster-Carr, including ¼ in. acrylicsheets for the tent base; self-adhesiverubber grommet strips (Reg. 8#93085-K12); centimeter-thick acrylicplates; weighting bags filled with leadshot or sand; high-vacuum grease(100 g tube, #1326K59); and parts forthe frame and lifting device, i.e., alu-minum pipe and pipe fittings (or usePVC pipe).

Barrier Films

Filmpak 1193 (oxygen transmission0.1 cm3 m - 2 day-1, 0.12 mmthickness)

Manufacturer

Ludlow Laminating & Coating Division4058 Highway 79Homer, LA 71040Tel: (318) 927-2531

Suppliers

CalTex Plastics2380 East 51st StreetVernon, CA 90056Tel: (213) 583-4140

Edco Supply327 36th StreetBrooklyn, NY 11232Tel: (718) 788-8108

Sealpak Flexible Packaging13826 S. Prairie AvenueHawthorne, CA 90250Tel: (310) 973-1321

Shannon Packaging Co.575 E. Edna PlaceCovina, CA 91723Tel: (818) 967-7329

Aclar Composites (Aclaroxygen transmission51.15 cm3 m - 2 day-1 calculated value,0.11 mm thickness)

Manufacturers

AclarAllied-Signal, Inc.P.O. Box 233RMorristown, NJ 07960Tel: (201) 455-2000

Film-O-Rap 7750 (FR 7750)Bell Fibre ProductsP.O. Box 1158Columbus, GA 31993Tel: (706) 323-7316

Filmpak 1177Ludlow Laminating & Coating DivisionHomer, LA 71040Tel: (318) 927-2531

82Appendix B

Suppliers

Keepsafe Systems, Inc.59 Glenmount Park RoadToronto, ON M4E 2N1CanadaTel: (416) 691-8854

Sealpak Flexible Packaging13826 S. Prairie AvenueHawthorne, CA 90250Tel: (310) 973-1321


George B. Woodcock & Co.21900 Marilla StreetP.O. Box 3009Chatsworth, CA 91313Tel: (818) 998-3774

Marvelseal 360 aluminized polymerfilm (oxygen transmission0.01 cm3 m-2 day-1, 0.13 mmthickness)

Manufacturer

Ludlow Laminating & Coating Division4058 Highway 79Homer, LA 71040Tel: (318) 927-2531

Suppliers

CalTex Plastics2380 East 51st StreetVernon, CA 90056Tel: (213) 583-4140


George B. Woodcock & Co.21900 Marilla StreetP.O. Box 3009Chatsworth, CA 91313Tel: (818) 998-3774

Materials and Suppliers 83

Preservation Equipment, Ltd.Shelfanger, DissNorfolk IP22 2DGEnglandTel: (44) 1379 651 527

Conservation by Design, Ltd.Timecare Works60 Park Road WestBedford MK41 7SLUnited KingdomTel: (44) 1234 217258

Multipak B.V.Industrieweg 143881 LB PuttenHollandTel: (31) 341 357474

BDF 200 (oxygen transmission4 cm3 m - 2 day-1, 0.025 mmthickness)

Supplier


Carbon Dioxide DetectionSystems

Suppliers

Bruel and KjaerDiv. of Spectris Technologies, Inc.2364 Park Central Blvd.Decatur, GA 30035Tel: (800) 332-2040

Campbell Scientific Inc.815 West 1800 NorthLogan, UT 84321Tel: (801) 753-2342

CEA Instruments, Inc.16 Chestnut StreetEmerson, NJ 07630Tel: (201) 967-5660

Gas Tech, Inc.8407 Central AvenueNewark, CA 94560Tel: (510) 745-8700

Teledyne Analytical Instruments16830 Chestnut StreetCity of Industry, CA 91740Tel: (818) 961-9221

Electrical Feed-Through,Hermetically Sealed Connector

Suppliers

Douglas Engineering Co.14 Beach StreetRockaway, NJ 07866Tel: (201) 627-8230

Omega Engineering1 Omega DriveStamford, CT 06907Tel: (203) 359-1660

Pave Technology Co., Inc.2751 Thunderhawk CourtDayton, OH 45414Tel: (513) 890-1100

Foiltec GmbHSteinacker 328717 BremenGermanyTel: (49) 421 69351-0(This company supplies valve that canbe converted.)

Heat Sealers

Suppliers

Tacking Iron (Sealector II)Conservation MaterialsP.O. Box 2884Sparks, NV 89432Tel: (702) 331-0582

McMaster-Carr Supply Co.9630 Norwalk Blvd.Santa Fe Springs, CA 90670Tel: (310) 692-5911

Spatula Tacking Iron (Tacking Tool74535AZ)McMaster-Carr Supply Co.9630 Norwalk Blvd.Santa Fe Springs, CA 90670Tel: (310) 692-5911

Appendix B 84

Spade-Tip Tacking Iron (Adem C6)Conservation MaterialsP.O. Box 2884Sparks, NV 89432Tel: (702) 331-0582

Pouch Sealing MachinePreservation Equipment, Ltd.Shelfanger, DissNorfolk 1P22 2DGEnglandTel: (44) 1379 651 527

Automatic Impulse Sealer(#92-96348)National Bag Co.2233 Old Mill RoadHudson, OH 44236Tel: (216) 425-2600

Humidification Systems(Commercial)

Suppliers

Edge Tech Moisture and HumiditySystems455 Fortune Blvd.Milford, MA 01757Tel: (800) 276-3729Fax:(800) 634-3010

General Eastern20 Commerce WayWoburn, MA 01801Tel: (617) 938-7070

Maheu & Maheu, Inc.710 Rue Bouvier, Suite 195Quebec, QC G2J 1C2CanadaTel: (418) 623 8000

Vacudyne Inc.375 Joe Orr RoadChicago Heights, IL 60411Tel: (708) 757-5200

Kojima Seisakujo Inc.6F, Popura BuildingOhdenma-cho, Nihonbashi, Chuo-kuTokyo 103Japan

Power-Plastics, Ltd.Station RoadThirskNorth Yorkshire YO7 1PZEnglandTel: (44) 1845 525503

Humidification Systems(Parts for Custom Systems)

Suppliers

Heavy-Duty Polypropylene Bottle(4 l volume, #02-923-5)Fisher Scientific (Headquarters)711 Forbes AvenuePittsburgh, PA 15219Tel: (412) 562-8300

FlowmeterCole-Parmer Instrument Co.7425 North Oak Park AvenueNiles, IL 60714Tel: (800) 323-4340

Nylon TubingCole-Parmer Instrument Co.7425 North Oak Park AvenueNiles, IL 60714Tel: (800) 323-4340

Swagelok Tube Fitting and ValvesAngeles Valve & Fitting Co.427 S. Victory Blvd.Burbank, CA 91502Tel: (213) 849-6257

B.E.S.T. Ventil & FittingGmbH FrankfurtRobert-Bosch-Str. 2063477 Maintal-DoernigheimGermanyTel: (49) 6181 46041

Birmingham Valve & Fitting Co., Ltd.Claymore, Tame Valley IndustrialEstateTamworthStaffordshire B77 5DQEnglandTel: (44) 827 281118

Humidity Monitor

Testo 610 Thermohygrometer

Supplier

Conservation Support SystemsEM-12105P.O. Box 91746Santa Barbara, CA 93190-1746Tel: (805) 682-9843

Manufacturers

ACR Systems, Inc.SurryBritish Columbia, CanadaTel: (604) 591-1128

Cole-Parmer Instrument Co.Vernon Hills, ILTel: (847) 549-7600

General Eastern InstrumentsWoburn, MATel: (617) 938-7070

Hy-Cal EngineeringEl Monte, CATel: (818) 444-4000

Hygrometrix, Inc.Alpine, CATel: (619) 659-9292

Kahn Instruments, Inc.Wethersfield, CTTel: (860) 529-8643

Metrosonics, Inc.Rochester, NYTel: (716) 334-7300

Omega EngineeringStamford, CTTel: (203) 359-1660

Rotronic Instrument Corp.Huntington, NYTel: (516) 427-3898

Teledyne Analytical InstrumentsCity of Industry, CATel: (818) 961-9221

TSI Inc.St. Paul, MNTel: (612) 483-0900


Vaisala, Inc., Sensors Systems Div.Woburn, MATel: (617) 933-4500

Leak Detectors

Manufacturers

Dresser Industries, Inc.Newton, CTTel: (203) 426-3115

Foxboro ICTSan Jose, CATel: (408) 432-1860

Gas Tech, Inc.Newark, CATel: (510) 745-8700

Matheson Gas ProductsMontgomeryville, PATel: (215) 641-2700

Neotronics Scientific, Inc.Flowery Branch, GATel: (770) 967-2196

Scott Specialty Gases Inc.Plumsteadville, PATel: (215) 766-8861

TIF Instruments, Inc.9101 NW 7th AvenueMiami, FL 33150Tel: (800) 327-5060Tel: (305) 757-8811Fax: (305) 757-3105

TSI Inc.St. Paul, MNTel: (612) 483-0900

Universal Sensors and Devices, Inc.Chatsworth, CATel: (818) 998-7121

Weed InstrumentsRound Rock, TXTel: (512) 434-2900

Nitrogen(High-Pressure Cylinders)

In addition to the primary suppliersbelow, others may be listed in localtelephone directories. Request guaran-teed analyses and price quotations forhigh-purity nitrogen or argon.

Suppliers

Gilmore Cryogenics9503 East Rush StreetSouth El Monte, CA 91733Tel: (213) 283-4721

Union Carbide Industrial Gases Inc.Linde Division200 Cottontail LaneSommerset, NJ 08875Tel: (201) 271-2600

Air ProductsUnited KingdomTel: (44) 01 256 7062

Linde AGSeitner Str. 7082049 HoelriegelskreuthGermanyTel: (49) 89 7446 1360

Nitrogen (Liquid-Nitrogen Tanks)(160 liters of liquid nitrogen,99,000 liters of gas, 99.998%purity)

Suppliers





Nitrogen Generator

Permea Prism Model 1300 (perfor-mance: 41 m3 day-1 of 99.99%nitrogen)

Suppliers

Air Products and Chemicals, Inc.7201 Hamilton Blvd.Allentown, PA 18195Tel: (610) 481-4911

Kojima Seisakujo Inc.6F, Popura BuildingOhdenma-cho, Nihonbashi, Chuo-kuTokyo 103Japan

Nitrogen Regulators(for Cylinders andLiquid-Nitrogen Tanks)

Suppliers






86Appendix B


Oxygen Absorbers

Ageless Z

Manufacturers

Mitsubishi Gas Chemical America, Inc.520 Madison Avenue25th FloorNew York, NY 10022Tel: (212) 752-4620

Mitsubishi Gas Chemical EuropeGmbHImmermannstrasse 454000 Düsseldorf 1GermanyTel: (49) 211 363080

Suppliers

Conservation Support SystemsP.O. Box 91746Santa Barbara, CA 93190-1746Tel: (805) 682-9843

Cryovac DivisionW. R. Grace & Co.16201 Commerce WayCerritos, CA 90701Tel: (562) 926-0424

Cryovac DivisionW. R. Grace & Co.P.O. Box 464Duncan, SC 29334Tel: (803) 433-3121

Keepsafe Systems, Inc.59 Glenmount Park RoadToronto, ON M4E 2N1CanadaTel: (416) 691-8854


CryovacGrace GmbHErlengang 3122841 NorderstedtGermanyTel: (49) 40 52601-0

Atco

Manufacturer

Standa Industrie184 Rue Maréchal-Gallieni14050 CaenFranceTel: (33) 31745489

Supplier

Preservation Equipment, Ltd.Shelfanger, DissNorfolk IP22 2DGEnglandTel: (44) 1379651527

Oxygen Analyzer(Specification Unknown)

Supplier


Oxygen Indicators (Passive)

Ageless-Eye

Manufacturers

Mitsubishi Gas Chemical America, Inc.520 Madison Avenue25th FloorNew York, NY 10022Tel: (212) 752-4620

Mitsubishi Gas Chemical EuropeGmbHImmermannstrasse 454000 Düsseldorf 1GermanyTel: (49) 211 363080

Suppliers


Appendix B 88

CryovacGrace GmbHErlengang 3122841 NorderstedtGermanyTel: 49 (0) 40 52601-0


Oxygen Indicator Eye

Supplier

Preservation Equipment, Ltd.Shelfanger, DissNorfolk IP22 2DGEnglandTel: (44) 1379 651 527

Oxygen Sensors (Electronic)

Manufacturers

Ametek, Inc., U.S. Gauge Div.Feasterville, PATel: (215) 355-6900

Control Instruments Corp.Westbury, NYTel: (516) 876-8400

Delta F. Corp.Woburn, WATel: (617) 935-4600

Gas Tech, Inc.Newark, CATel: (510) 745-8700

Matheson Gas ProductsMontgomeryville, PATel: (214) 641-2700

Neutronics Inc.Exton, PATel: (610) 524-8800

Panametrics Inc.Process Control Instrument DivisionTel: (800) 833-9438

BühlerMess- u. RegeltechnikGmbH & Co. KG40880 RatingenGermanyTel: (49) 2102 49890

Teledyne, Ltd.Southhall LaLondon VB2 5NHEnglandTel: (44) 181 5719596

Temperature Control Systems

Suppliers

Honeywell IndustrialAutomation & Control16404 N. Black Canyon HighwayPhoenix, AZ 85023Tel: (602) 863-5000

Johnson Controls Inc.Milwaukee, WlTel: (800) 333-2222

Omega Engineering1 Omega DriveStamford, CT 06907Tel: (203) 359-1660

Yokogawa Corporation of AmericaTest & Measurement Division2 Dart RoadNewnan, GA 30265Tel: (800) 258-2552

Temperature Monitors

Manufacturers

ACR Systems, Inc.Surry, British Columbia, CanadaTel: (604) 591-1128

Cole-Parmer Instrument, Co.Vernon Hills, ILTel: (847) 549-7600

Everest Interscience, Inc.Tucson, AZTel: (520) 797-0927

Omega EngineeringStamford, CTTel: (203) 359-1660

R. M. Young Co.Traverse City, MlTel: (616) 946-3980

Raytek, Inc.Santa Cruz, CATel: (408) 458-1175

Vaisala, Inc., Sensors Systems Div.Woburn, WATel: (617) 933-4500

Weed InstrumentRound Rock, TXTel: (512) 434-2900

Young Environments SystemsRichmond, British Columbia, CanadaTel: (604) 276-9923

YSI Inc.Yellow Springs, OHTel: (513) 767-7241

Treatment Containers

Rigid Enclosures

Suppliers

Bemco Inc.Environmental & Space SimulationSystems2255 Union PlaceSimi Valley, CA 93065Tel: (805) 583-4970

Harris Environmental Systems, Inc.11 Connector RoadAndover, MA 01810Tel: (508) 475-0104

Thermatron Industries291 Kollen Park DriveHolland, Ml 49423Tel: (616) 393-4580

Vacudyne Inc.375 Joe Orr RoadChicago Heights, IL 60411Tel: (708) 757-5200(Company supplies conversion kitfor chambers previously used forethylene oxide fumigation.)

Soft Enclosures(Rentokil Bubble)

Suppliers

Maheu & Maheu, Inc.710 Rue Bouvier, Suite 195Quebec, QC G2J1C2Canada



90Materials and Suppliers

91

Appendix C

Common andScientific Namesof Insect Species

Common Name Scientific Name

American cockroach

Australian spider beetle

Black carpet beetleBook beetle

Book louseBrownbanded cockroach

Cabinet beetleCasemaking clothes moth

Cigarette beetleCommon clothes moth

Confused flour beetle

Deathwatch beetle

Drugstore beetle

Drywood termite

FirebratFruit flyFur beetle

Furniture beetle

Furniture carpet beetleGerman cockroach

Grain mite

Hide beetle

Khapra beetleLarder beetleLonghorn borer beetleªMerchant grain beetleMold mite

Odd beetleOld house borer ªPowderpost beetle

Red flour beetle

Redlegged ham beetleRice weevilSilverfish

Varied carpet beetleWebbing clothes moth

a One of several common names for Hylotrupes bajulus.

Periplaneta americana

Ptinus ocellus

Attagenus piceus

Nicobeum casteneum

Liposcelis corrodens

Supella longipalpa

Trogoderma inclusum

Tinea pellionella

Lasioderma serricorne


Tribolium confusion

Xestobium rufovillosum

Stegobium paniceum

Cryptotermes brevis

Thermobia domestica

Drisopholus melanogaster

Attagenus fasciatus

Anobium punctatum

Anthrenus flavipes

Blattella germanica

Acarus siro

Dermestes maculatus

Trogoderma granarium

Dermistes lardarius

Hylotrupes bajulus

Oryzaephilus mercator

Tyrophagus putrescentia

Thylodrias contractus

Hylotrupes bajulus

Lyctus spp.

Tribolium castaneum

Necrobia rufipes

Sitophilus oryzae

Lepisma saccharina

Anthrenus verbasci


93

References AliNiazee, M. T.1972Susceptibility of the confused and red flour beetles to anoxia produced byhelium and nitrogen at various temperatures. Journal of Economic Ento-mology 65:60.

Annis, P. C.1987Towards rational controlled atmosphere dosage schedules: A review of currentknowledge. In Proceedings of the 4th Working Conference on Stored-ProductProtection, Tel Aviv, Israel, September 21-26, 1986, ed. E. Donahaye andS. Navarro, 128-48. Bet Dagen, Israel: Agricultural Research Organization.

Arai, H.r and H. Mori1980Studies on the long term conservation of cultural property, using biaxially ori-ented poly(vinyl alcohol) film bags to prevent the biodeterioration of culturalproperties. Kobunkazai no Kagaku 25:89-102.

Arney, J. S., A. J. Jacobs, and R. Newman1979Influence of oxygen on the fading of organic colourants. Journal of the Ameri-can Institute for Conservation 18:108-17.

Bailey, C. H.r and A. M. Gurjar1918Respiration in stored wheat. Journal of Agricultural Research 12:685.

Banks, H.J.1979Recent advances in the use of modified atmospheres for stored product pestcontrol. In Proceedings of the 2nd International Working Conference on Stored-Product Entomology, September 10-16, 1978, Ibadan, Nigeria, 198-217. Savan-nah, Ga.: n.p.

Banks, H. J., and P. C. Annis1977Suggested procedures for controlled atmosphere storage of dry grain. Common-wealth Scientific and Industrial Research Organization (CSIRO) Division of Ento-mology Technical Paper, 13.

1990Comparative advantages of high CO2 and low O2 types of controlled atmo-spheres for grain storage. In Food Preservation by Modified Atmospheres, ed.M. Calderon and R. Barkai-Golan, 93-122. Boston: CRC Press.

Beauchamp, R., J. Kingsolver, and T. Parker1981Illustrated Guide to Common Insect Pests in Museums: A Status Report (1980).Lawrence, Kans.: Association of Systems Collections.

Binker, G.1993aNew concepts for environment protection and new developments in the fumiga-tion of cultural property. In Fumigation as a Means of Wood Pest Control: Pro-ceedings of a Conference Held by the Restoration Studios of the Bavarian StateConservation Office in Munich on October 22, 1993, 90-98. Arbeitshefte desBayerischen Landesamtes für Denkmalpflege, no. 75. Munich: Bayerisches Land-esamt für Denkmalpflege.

1993bMit Kohlendioxid gegen Insektbefalls: Wie kann die Einwirkzeit verkürztwerden? Restauro 4:222.

1996Communication with the author.

Blanchette, R. A.1995Biodeterioration of archaeological wood. Biodeterioration Abstracts9(2):113-27.

Buck, Richard D.1952A note on the effect of age on the hygroscopic behavior of wood. Studies inConservation 1:39-44.

Burke, J.1992Vapor barrier films. WAAC (Western Association of Art Conservation) Nevvs-letter 14(2):13-17.

Bursell, E.1957Spiracular control of water loss in the tsetse fly. Proceedings of the Royal SocietyLondon A(32):21-29.

Calderon, M.r and R. Barkai-Golan1990Food Preservation by Modified Atmospheres. Boston: CRC Press.

Carlson, S. D.1995Measuring the metabolism of small organisms. Scientific American(December):102-3.

Child, B.1994The Thermo Lignum process for insect pest control. Paper ConservationNews (9): 72.

Child, R. E.1988Fumigation: A new direction? Conservation Today: Papers Presented at theUKIC 30th Anniversary Conference, vol. C-7, ed. Victoria Todd. London: UnitedKingdom Institute of Conservation.

Cipera, L, and M. Segal1996The use of carbon dioxide to control insects at the Canadian Museum of Civiliza-tion, Hull. In International Institute for Conservation—Canadian Group, 22ndAnnual Conference, Montreal, 31-41. Ottawa, Canada: IIC—CG.

Daniel, V., G. Hanlon, and S. Maekawa1993Eradication of insect pests in museums using nitrogen. WAAC (Western Associa-tion for Art Conservation) Newsletter 3:15-19.

References 94

References 95

Davis, R.r and E. G. Jay1983An overview of modified atmospheres for insect control. Association of Opera-tive Millers—Bulletin, 4026-29.

deCesare, K. B. J.1990Safe nontoxic pest control for books. Abbey Newsletter 14(1): 16.

Dendy, A.1918Report on the effect of airtight storage upon grain insects. II. Reports of theGrain Pests (War) Committee, Royal Society of London 3:3.

Donahaye, E.1990Laboratory selections of resistance of the red flour beetle, Tribolium castaneum(Herbst), to an atmosphere of low oxygen concentration. Phytoparasitica18:189-202.

Edney, E. B.1967Water balance in desert arthropods. Science 156:1056.

Elect, K.1996Communication with the author.

Elect, K., and S. Maekawa1997Rentokil bubble in nitrogen anoxia treatment of museum pests. Studies in Con-servation 42:247-52.

Ertelt, P.1993Studies on controlled thermal treatment of pest infested wood. Diploma thesis,Department of Wood Technology, Fachhochschule Rosenheim, Germany.

Gilberg, M.1989aInert atmosphere fumigation of museum objects. Studies in Conserva-tion 34:80-84.

1989bAgeless oxygen scavenger. Australian Institute for Conservation of CulturalMaterial National Newsletter 32.

1990Inert atmospheres disinfestation using Ageless oxygen scavenger. In ICOMCommittee for Conservation 9th Triennial Meeting, Dresden, German Demo-cratic Republic, 26-31 August 1990, Preprints, vol. 2, ed. K. Grimstad, 812-16.Los Angeles: ICOM Committee for Conservation.

1991The effects of low oxygen atmospheres on museum pests. Studies in Conserva-tion 36:93-98.

References 96

Gilberg, M., and A. Roach1992Inert atmosphere disinfestation of museum objects using Ageless oxygen ab-sorber. In Biodeterioration of Cultural Property 2: Proceedings of the 2nd Inter-national Conference on Biodeterioration of Cultural Property, 397-406. Yoko-hama: International Congress of Biodeterioration of Cultural Properties.

Hanlon, G., V. Daniel, N. Ravenel, and S. Maekawa1992Dynamic system for nitrogen anoxia of large museum objects: A pest eradicationcase study. In Biodeterioration of Cultural Property 2: Proceedings of the 2ndInternational Conference on Biodeterioration of Cultural Property, 387-96.Yokohama: International Congress of Biodeterioration of Cultural Properties.

Hassan, A. A. G.1943The structure and mechanism of the spiracular regulatory apparatus in adultDiptera and certain other groups of insects. Royal Entomological Society ofLondon 94:103-53.

Inoue, Y., H. Hatakeyama, and I. Yoshino1994Oxygen indicator. U.S. Patent 5,358,876. (25 October).

Jacobs, J.1996Communication with the author.

Jay, E. G.1984Imperfections in our current knowledge of insect biology as related to theirresponse to controlled atmospheres. In Controlled Atmospheres and Fumigationin Grain Storage, ed. B. E. Ripp. Amsterdam: Elsevier.

Jay, E. G., R. T. Arbogast, and G. C. Pearman Jr.1971Relative humidity, its importance in the control of stored-product insects withmodified atmospheric gas concentrations. Journal of Stored ProductsResearch 6:325.

Jay, E. G., and W. Cuff1981Weight loss and mortality of three life stages of Tribolium castaneum (Herbst)when exposed to modified atmospheres. Journal of Stored ProductsResearch 17:117.

Johnston, A.1996Communication with the author.

Kamba, N.1994Performance of wooden storage cases in regulation of relative humidity change.In Preventive Conservation: Practice, Theory and Research, ed. Ashok Roy andPerry Smith, 181-84. London: International Institute for Conservation of Historicand Artistic Works.

Koestler, R.J.1992Practical application of nitrogen and argon fumigation procedures for insect con-trol in museum objects. In Biodeterioration of Cultural Property 2: Proceedingsof the 2nd International Conference on Biodeterioration of Cultural Property,94-96. Yokohama: International Congress of Biodeterioration of CulturalProperties.

Koestler, R. J., and T. F. Mathews1994Application of anoxic treatment for insect control in manuscripts of the library ofMegisti Laura, Mount Athos, Greece. Actes des Deuxièmes Journées Interna-tional d’Etudes de l’ARSAG, Paris, 59-62.

Koestler, R. J., E. Parreira, E. Santoro, and P. Noble1993Visual effects of selected biocides on easel painting materials. Studies in Conser-vation 38:265-73.

Lambert, F. L., V. Daniel, and F. D. Preusser1992The rate of absorption of oxygen by Ageless; the utility of an oxygen scavengerin sealed cases. Studies in Conservation 37:267-74.

Lambert, F., and S. MaekawaForthcomingEvaluation of Oxygen Exchange Rates.

Maekawa, S., and K. Elert1996Large-scale disinfestation of museum objects using nitrogen anoxia. In ICOMCommittee for Conservation 11th Triennial Meeting, Edinburgh, Scotland,1-6 September 1996, Preprints, vol. 1, ed. J. Bridgland, 48-53. London:James and James.

Maekawa, S., F. D. Preusser, and F. L. Lambert1993An hermetically sealed display and storage case for sensitive organic objects ininert atmospheres. In Biodeterioration of Cultural Property 2: Proceedings ofthe 2nd International Conference on Biodeterioration of Cultural Property,374-85. Yokohama: International Congress of Biodeterioration of CulturalProperties.

Mellanby, K.1934The site of water loss from insects. Proceedings of the Royal Society LondonB(116):139-49.

Michalski, S.1993Relative humidity: A discussion of correct/incorrect values. In ICOM Committeefor Conservaton 10th Triennial Meeting, Washington D.C., USA, 22-27 August1993, Preprints, vol. 1, ed. J. Bridgland, 22-27. Paris: ICOM Committee forConservation.

References 97

References 98

Muchnic, S.1993Kienholz vs. Mothra: Artist's Dodge beats the censors, but then came theinsects. Los Angeles Times, 19 May, F1, F4, and F5.

Navarro, S.1978The effects of low oxygen tensions on three stored-product insect pests. Phyto-parasitica 6:51.

Newton, J.1990Carbon dioxide fumigation in a heated portable enclosure. In Proceedings of theFifth International Working Conference on Stored-Product Production, ed.F. Fleurat-Lessard and P. Ducom, vol. 2, 877-85. Bordeaux, France: Impri-merie du Médoc.

Newton, J., M. Abey-Koch, and D. B. Pinniger1996Controlled atmosphere treatment of textile pests in antique curtains using nitro-gen anoxia—A case study. In Proceedings of the 2nd International Conferenceon Insect Pests in the Urban Environment, Heriot-Watt University, Edinburgh,Scotland, ed. K. B. Wildey. (Applied Technology Session.) [London?]: n.p.

Noble-Nesbitt, J.1989Spiracular closing membranes in the firebrat, Thermobia domestica (Packard)(Thysannra). Tissue and Cell 21 (1):93-99.

Pine, S., V. Daniel, and S. Maekawa1993Large scale fumigation of organic materials using nitrogen. In Papers Presentedat the Wooden Artifacts Croup Specialty Session, June 4, 1993, AIC AnnualMeeting, Denver, Colorado. Washington, D.C.: AIC Wooden Artifacts Group.

Reichmuth, C.1987Low oxygen content to control stored product insects. In Proceedings of the 4thWorking Conference on Stored-Product Protection, Tel Aviv, Israel, September21-26, 1986, ed. E. Donahaye and S. Navarro, 194-207. Bet Dagen, Israel: Agri-cultural Research Organization.

Reichmuth, C., A. Unger, W. Unger, G. Blasum, H. Piening, P. Ronde-Henr,and R. Plarre1993Nitrogen flow fumigation for the preservation of wood, textiles, and otherorganic material from insect damage. In International Conference on ControlledAtmospheres and Fumigation in Grain Storage, June 1992, Winnipeg, Canada,121-28. Jerusalem: Caspit Press.

Rust, M., and J. Kennedy1993The feasibility of using modified atmospheres to control insect pests in muse-ums. Internal report, The Getty Conservation Institute Scientific Program.

1995Pest control by oxygen deprivation using nitrogen. Internal report, The GettyConservation Institute Scientific Program.

Rust, M. K., V. Daniel, J. R. Druzik, and F. D. Preusser1996The feasibility of using modified atmospheres to control insect pests in muse-ums. Restaurator 17:43-60.

Smith, C. P.1995Controlled atmosphere treatment, Marseilles. Internal report, Rentokil Company,East Grinstead, West Sussex, United Kingdom.

1996Communication with the author.

Smith, C. P., and J. Newton1991Carbon dioxide: The fumigant of the future. In International Seminar onResearch in Preservation and Conservation, Columbia University, N.Y., ed.K. G. Saur. New York: International Federation of Library Associations.

Story, K. O.1985Approaches to Pest Management in Museums. Suitland, Md.: ConservationAnalytical Laboratory, Smithsonian Institution.

Strang, T.1992A review of published temperatures for the control of pest insects in museums.Collection Forum 8(2):41-67.

1995The effect of thermal methods of pest control on museum collections. In Biode-terioration of Cultural Property 3: Proceedings of the 3rd International Confer-ence on Biodeterioration of Cultural Property, July 4-7, 1995, 199-218, ed.C. Aranyanak and C. Singhasiri. Bangkok: Conservation Science Division, Officeof Archaeology and National Museums, The Fine Arts Department.

Thomson, R. S.1995The effect of the Thermo-Lignum pest eradication treatment on leather andother skin products. Report, The Leather Conservation Center, Northampton,United Kingdom.

Unger, A., W. Unger, and C. Reichmuth1992The fumigation of insect-infested wood sculptures and paintings with nitrogen.In Biodeterioration of Cultural Property 2: Proceedings of the 2nd InternationalConference on Biodeterioration of Cultural Property, 440-46. Yokohama: Inter-national Congress of Biodeterioration of Cultural Properties.

Valentín, N.1990Insect eradication in museums and archives by oxygen replacement, a pilotproject. In ICOM Committee for Conservation 9th Triennial Meeting, Dresden,German Democratic Republic, 26-31 August 1990, Preprints, vol. 2, ed.K. Grimstad, 821-23. Los Angeles: ICOM Committee for Conservation.

References 99

References 100

1993Comparative analysis of insect control by nitrogen, argon, and carbon dioxide inmuseum, archive, and herbarium collections. International Biodeterioration andBiodegradation 32:263-78.

Valentín, N., and F. Preusser1989Nitrogen for biodeterioration control on museum collections. In The Third Meet-ing of the Pan American Biodeterioration Society, Washington.

Warren, S.1996The carbon dioxide fumigation program at the National Museum of Science andTechnology, Ottawa. In International Institute for Conservation—CanadianGroup, 22nd Annual Conference, Montreal, 53-60. Ottawa, Canada: IIC—CG.

Wiggles worth, V. B., and G. D. Gillett1936The loss of water during ecdysis in Rhodnius prolixus Stal (Hemiptera). In Pro-ceedings of the Royal Society London A(11):104-7.

Zebe, E.1991Arthropods; Insecta. In Metazoan Life without Oxygen, ed. C. Bryant. NewYork: Chapman Hall.

Index Note: The index designations f and t refer, respectively, to figures and tables.

AAbey-Koch, M., with Newton, 45

ABS. See acrylonitrile-butadiene-styrene

Aclar, 19, 19t, 49, 50-51

suppliers of, 82

ACR Systems, Inc., 37

acrylonitrile-butadiene-styrene (ABS), 74

Adem C6, 25t

adhesive strips, 26

Ageless, 8, 22, 23-24, 35-36

Burke's use of, 42-43

color change in, 35

durability of, 35-36

forms of, 22-23

intensity of, 23

suppliers of, 86-87

type C, 35

type K, 35

Ageless Z, 22, 23

Ageless Z-2000, 46

Air Products and Chemicals Corporation, 21,

58

AliNiazee, M. T., 4, 5

Altarion, 76

aluminized polyethylene, 18

aluminum, 19t, 42

American cockroach, kill time for, 12t, 69t

anaerobic respiration, 5

Annis, P. C., 8

with Banks, 8, 9, 40, 67, 68

Anobiidae, 9

anoxia, 7-15

in barrier-film bags, 41-47

general procedures for, 41-42

dynamic method, 17, 49-55

in constructed containments, 50-55

at J. Paul Getty Museum, 50-52

in small pouches, 49-50

early studies in, 8-9

flexible chambers for, 57-62

at high humidity, 4-5

insects resistant to, 5

materials for, 17-27

methods for, 17-27

reusable systems, 57-66

rigid chambers for, 62-66

static method, 17

101

static-dynamic method, 17

containment systems in, 17-18

thermal techniques, 13-14

anoxia systems, suppliers of, 81

anoxia tents, suppliers of, 81

Approaches to Pest Management in Museums

(Story), 7

Arai, H., 7

Arbogast, R. T., with Jay, 1, 2, 4

argon, 4, 15t, 31, 32f

cost of, 21t, 33

density of, 34t

grades of, 20

and kill times, 13

versus nitrogen, 33-34

radius of, 4

safety issues with, 38-39

suppliers of, 20

and temperature, 3f

Arney, J. S., xi

Atco 3000, 46

Atco LH, 22-23

Atco SA, 22

Atco, suppliers of, 87

Australian Museum, 7-8, 22, 23-24

Australian spider beetle, kill time for, 69t

automatic impulse sealer, 24, 25t

azine dye, oxygen-related color change in, 35,

35f

BB & G Mini Bubble. See Rentokil bubble, large

Back Seat Dodge '38, treatment of, 53-54, 54f

bacteria, and nitrogen anoxia, 9

Bailey, C. H., 7

Banks, H. J., 8, 9, 40, 67, 68

bar sealer, 24, 25t

Barkai-Golan, R., with Calderon, 7

barrier-film bags, 18-20

anoxia in, 41-47

composition of, 19t, 24

conservators' experience with, 42-47

cost of, 20t

on large scale, 46-47, 47f

making by hand, 41

making window in, 41-42

suppliers of, 82

Beauchamp, R., 10

Bedfont Scientific Instrument BT 2000, 46

Biebel, W., contact information, 80

Binker, G., 76, 77, 78

contact information, 80

black carpet beetle, 12

burrowed, 15t

kill time for, 12t

Blasum, G., with Reichmuth, 49

Bone, L., 49

book beetle, 12

burrowed, 15t

kill time for, 12t

book lice

and argon anoxia, 33

kill times for, 69t

books, and argon, 15t

boring beetles, 67

brownbanded cockroach, kill times for, 12t

Bruel and Kjaer Instruments Inc., 36

Buck, R. D., 31

Burke, J., xi, 18, 20, 26, 42-43, 72, 73


burrowed insects, 14-15

Bursell, E., 1

cabinet beetle, kill times for, 12t

Calderon, M., 7

Canada, carbon dioxide fumigation in, 74-76

Canadian Conservation Institute, 43, 74

Canadian Museum of Civilization, Quebec,

75-76

carbon dioxide, 7

cost of, 21t

density of, 34t

effectiveness of, versus nitrogen, 67

and kill time, 11

safety issues with, 38-40, 40t

carbon dioxide detection systems, suppliers of,

83

carbon dioxide exchange, 1-2

carbon dioxide fumigation, 31-32, 32f, 67-78

in Canada, 74-76

duration of exposure, 68

in Germany, 76-78

kill times with, 68-70, 69t

in pressure chambers, 78

requirements for, 67-70

in United States, 70-74

carbon monoxide monitors, 36

Carlson, S. D., 38

Index

carpet beetles, 8

treatment of, 60

Casebolt, D., 49

casemaking clothes moths

kill times for, 45-46

treatment of, 60

Central Science Laboratories, 45

Child, B., 13

Child, R. E., 70

church, fumigation of, 76-78, 78f

cigarette beetles, 5, 8, 12, 24

burrowed, 15t

kill times for, 11, 12t, 13f, 69t

Cipera, L, 31, 68, 75

closures, 24-26

adhesive strips, 26

clamps, 25-26

heat sealers, 24-25

zippers, 25

Cole-Parmer Instrument Company, 34

Cole-Parmer minihygrothermographs, 30

Colton, S., 53, 65


Commonwealth Scientific and Industrial

Research Organization, 8

Compressed Gas Association, 38-39

confused flour beetles

and humidity, 2, 3f, 4f

kill times for, 5f, 12t

Conservation Support Systems, 26, 49

conservators, contact information, 79-80

console table, treatment of, 51

Cranach, Lucas the Younger, 50

Cuff, W., with Jay, 3

DDaniel, V., 17, 29, 53


with Hanlon, 17, 29

with Lambert, 22, 23

with Maekawa, 65

with Pine, 60

with Rust, 9

Davis, R., 68

deathwatch beetle, 12

burrowed, 15t

deCesare, K. B. J., 7

dehydration, 1-2

and weight loss, 2

del Sarto, Andrea, 17

Delta F. Corporation, 34

102

Dendy, A., 7

dermestids, 5, 9

dessication, 1-4

display cases, nitrogen-filled, 8-9

documents, and argon, 15t

Donahaye, E., 2, 4

drugstore beetles, 8, 12, 24

burrowed, 15t

kill times for, 12t

Druzik, J. R., with Rust, 9

drywood termites

kill times for, 12t

in panel painting, 17

EEdge Tech Moisture and Humidity Systems, 37

Edney, E. B., 3

Egyptian Antiquities Organization, 8

electronic oxygen monitors, 34

Elert, K., 36, 59

with Maekawa, 21, 57

Ephestia cautella, and weight loss, 2, 3

Ertelt, P., 13

ethylene-vinyl acetate copolymer, 24

ethylene-vinyl alcohol copolymer, 19t

FFang, M., 43-44

Film-O-Rap, 19t, 44

Filmpack, 19t

Fine Arts Museums of San Francisco,

California, 49

fire screen, treatment of, 50-51

firebrat

and carbon dioxide, 2

kill times for, 12t

flexible chambers, for anoxia, 57-62

suppliers of, 89

flour beetle, and humidity, 3-4

food preservation, 7-8

frames, and argon, 15t

French Ministry of Culture, 46-47

Freon, 61

fruit fly, kill times for, 12t

fumigation bubble, in static-dynamic method,

17-18

fumigation, with carbon dioxide, 67-78

fumigators, contact information, 80

Fundación Bartolomé Pons Cabrera, Spain, 54

fungi, and nitrogen anoxia, 9

C

furniture beetles, 12

burrowed, 15t

kill times for, 12t, 13, 50

larvae, kill times for, 69t

treatment of, 60

furniture carpet beetles, 24

kill times for, 11, 12t, 13f

Ggas cylinders, 20-21

gas supplies, 20-21

cost of, 20

Gas Tech, Inc., 34, 36

gas-transport system, 1-2

General Eastern Instruments, 37

German cockroach, kill times for, 12t

Germany, carbon dioxide fumigation in, 76-78

Getty barrier-film tent, 57f, 57-59, 58f

Getty Conservation Institute, 7-8, 8-9, 21, 29,

53, 57, 65

Gilberg, M., 7-8, 9, 11, 12, 13, 22, 23, 35, 38,

42, 43, 49


Gillett, G. D., with Wigglesworth, 1

grain mite, kill times for, 69t

Great Lavra Monastery, 33

Gurjar, A. M., with Bailey, 7

HHampshire County Museum Council Museum

Service, United Kingdom, 63

Hanlon.G., 17, 29, 50-51, 53


with Daniel, 17, 29

Hardigg case, as anoxia chamber, 65, 65f

Hardigg Industries, 65

Hassan, A. A. G., 1

heat sealers, 24-25

suppliers of, 83-84

types of, 24-25, 25t

helium, 33

and kill time, 5f

radius of, 4

hermetic seal, 57-58, 58f

suppliers of, 83

hide beetle, kill times for, 69t

Hinerman, R., 24

homemade rigid chambers, 62-65, 63f

humidification, 29-32

decision against, 31

new module for, 30f

Index

suppliers of systems, 84

three-jar system, 29f, 29-30, 30f

humidity, 3-4

and anoxia, 4-5

and nitrogen, 4f

and temperature, 13-14

humidity monitors, suppliers of, 84-85

Huntington Library Art Collections and

Botanical Gardens, California, 66

HVAC monitoring industry, 37

Hy-Cal Engineering, 37

Hylotrupes bajulus, 12, 67

Iimpulse sealer, 24, 25t

inert gases

permeability of, 4

type of, and kill time, 5f

insect exposure chambers, 10, 11f

insect mortality, mechanisms of, 1-5

insects

common and scientific names of, 91

in historic collections, 12

in North American museums, 9, 10f

in North American natural history museums,

10t

Institute of Museum Services, 72

Instituto de Conservación y Restauración de

Bienes Culturales. See Instituto del

Patrimonio Histórico Español, Madrid

Instituto del Patrimonio Histórico Español,

Madrid, 44-45

Italian armchair, treatment of, 51f, 51-52

oxygen concentration during, 52, 52f

oxygen leakage during, 51, 52f

JJ. Paul Getty Museum, 29

dynamic anoxia at, 50-52

Jacobs, A. J., with Arney, xi

Jacobs, J., 68, 73


Jay, E.G., 1, 2, 3, 4, 67

with Davis, 68

Johnston, A., 63


KKamba, N., 31

Keepsafe, 19t

Kennedy, J., with Rust, 9, 10, 11, 12, 13, 15,

29, 38, 46, 50, 53, 70

103

khapra beetle, kill times for, 69t

Kienholz, Edward, 53

kill times

at all life stages, 12t

with carbon dioxide fumigation, 68-70, 69t

determination of, 9-13

Kingsolver, J., with Beauchamp, 10

Koestler, R. J., xi, 17, 33, 38, 44, 62

Kronkright, D., 59


LLa Coruña, Spain, 45

LACMA. See Los Angeles County Museum of

Art

Lambert, F. L., 14, 22, 23

with Maekawa, 9

larder beetle, kill times for, 12t

leak detectors, 37

suppliers of, 85

leaks, 44

in large Rentokil bubbles, 61, 62f

life signs, monitoring of, 37-38

liquid nitrogen, 21. See also nitrogen

cost of, 21t

longhorn borer beetle, 12

kill times for, 12t

Los Angeles County Museum of Art (LACMA),

44, 53, 65-66

Maekawa, S., 9, 21, 53, 57, 65


with Daniel, 17, 29

with Elert, 59

with Hanlon, 17, 29

with Lambert, 14

with Pine, 60

Maheu & Maheu, Inc., 70

Maheu, M., contact information, 80

March, Bartolomé, 54

Marvelseal, 18, 19t, 43-44

suppliers of, 82-83

Materials and Equipment for Anoxic

Fumigation, 43

materials, contact information, 81-89

Mathews, T. R, with Koestler, 33, 38

Mellanby, K., 1

Mendocino County Museum, California, 24,

43

merchant grain beetle, kill times for, 69t

M

methylene blue, oxygen-related color change

in, 35, 35f

Metropolitan Museum of Art, New York, 44

Michalski, S., 31

minihygrothermographs, 30

Mitchell Instrument Company, 34

Mitsubishi Gas Chemical Company, 8, 22, 26,

35

MNM. See Museum of New Mexico

mold mite, kill times for, 69t

monitors, 34-37

carbon monoxide, 36

electronic oxygen, 34

humidity, 36-37

suppliers of, 84-85

oxygen, 36

passive oxygen, 34-35

as remote sensors, 34

with sample aspirator, 34

temperature, suppliers of, 88-89

Mori, H., with Arai, 7

mortality, measurement of, 37-38

Mottisfont Abbey, United Kingdom, 45

Muchnic, S., 53

Museum of Fine Arts, Houston, 30, 60-62,

63-64

Museum of International Folk Art, Santa Fe,

59-60

Museum of New Mexico (MNM), 59

Museum of St. John, London, 70

NNational Museum of Science and Technology,

Ottawa, 30, 74-75

National Trust, 45

natural history museums, insect pests in, 10t

Navarro, S., 2, 3

needle valves, 29, 30

Newman, R., with Arney, xi

Newton, J., 45, 70


with Smith, 68, 70

nitrogen, 4, 9

versus argon, 33-34

and burrowed insects, 14t

versus carbon dioxide, 67

cost of, 21t, 33

density of, 34t

grades of, 20

and humidity, 4f

and kill time, 5f

Index

radius of, 4

with Rentokil bubbles, 59, 60-61

safety issues with, 38-39

suppliers of, 20, 85-86

and temperature, 3f

nitrogen flow fumigation, 49-50

nitrogen generator, suppliers of, 86

nitrogen purging, 17

nitrogen regulators, suppliers of, 86

Noble, P., with Koestler, xi

Noble-Nesbitt, J., 2

nontoxic fumigants, safe use of, 38-40

nylon, 24

nylon-6, 19, 19t

Oakland Museum, California, 18, 26, 42, 43,

72f, 72-73, 73f

odd beetles, 60, 68

Oddy test, 26

old house borer, 12

burrowed, 15t

Old Sturbridge Village, Massachusetts, 70, 71

Omega Engineering, Inc., 34

O-ring sealers, 26-27, 27f

Oryzaephilus, and nitrogen anoxia, 9

O-Seal Straight Thread Male Connectors, 26

oxygen

density of, 34t

and kill times, 13f

oxygen absorbers, 17

suppliers of, 86-87

oxygen analyzer, 46

suppliers of, 87

Teledyne, 10, 14, 44, 59

oxygen exchange, 1-2

oxygen indicators, suppliers of, 87-88

oxygen monitors, 36

oxygen scavengers, 17, 18, 22-24

intensity of, 23

oxygen sensors, suppliers of, 88

oxygen-deficient atmospheres, safety concerns

with, 38-40, 39t

Ppanels, treatment of, 17

with argon, 15t

paper, buffering effect of, 31, 32f

Parker, T., with Beauchamp, 10

Parreira, E., with Koestler, xi

passive oxygen monitors, 34-35

104

Pearman, G. C, Jr., with Jay, 1, 2, 4

Phoebe Apperson Hearst Museum of

Anthropology, California, 43-44

piano, and argon, 15t

Piening, H., with Reichmuth, 49

Pine, S., 60-62, 63-65


Pinniger, D., with Newton, 45

plants, and argon, 15t

Plarre, R., with Reichmuth, 49

plastic bags

for anoxia, characteristics of, 18, 19t

in static-dynamic method, 17-18

poly(chlorotrifluoroethylene), 19, 19t, 42, 57

polyethylene, 19t

aluminized, 18

polyethylene terephthalate, 19, 19t

polymer chemistry, 18-20, 19t

polypropylene, 19t

polypropylene laminate with aluminum, 19t

polypropylene water tank, as anoxia chamber,

63

polystyrene, 19t

polyvinyl chloride, 19, 19t

polyvinylidene chloride, 24

portals, 26-27

powderpost beetles, 8, 24

burrowed, 15, 15t

kill times for, 12t, 13, 50, 72

Power Plastics, 59, 70

pressure chambers, carbon dioxide fumigation

in, 78

Preusser, F. D.

with Lambert, 22, 23

with Maekawa, 9

with Rust, 9

with Valentín, N., 9, 12

Prism nitrogen generator, 21, 58

PVC pipe, as anoxia chamber, 63-64, 64f

PVDC, 19t

RRattigan, G., contact information, 80

Ravenel, N., with Hanlon, 17, 29

red flour beetles

and carbon dioxide, 2

and high humidity, 4-5

and humidity, 2, 3f, 4f

kill times for, 5f, 69t

and weight loss, 2, 3

O

Reichmuth, C, 49-50, 67

with Unger, 50

Reine de Gallicia, Spain, 45

relative humidity (RH), 2, 3f. See also humidity

relative humidity monitors, 36-37

Rentokil bubble

large, 60-62

small, 59-60

construction of, 59-60

suppliers of, 89

Rentokil, Inc., 45-46, 70

RH. See relative humidity

Rhysopertha, and nitrogen anoxia, 9

rice weevil, 8

kill times for, 12t, 69t

and nitrogen anoxia, 9

and weight loss, 2

rigid chambers

for anoxia, 62-66

homemade, 62-65, 63f

turnkey unit, 66

Vacufume, 65-66

suppliers of, 89

Roach, A., with Gilberg, 13, 23

Ronde-Henr, P., with Reichmuth, 49

Roosa, Mark, 66

Royal Mummies, 8

Royal Ontario Museum, Toronto, 65

Rust, M., 9, 10, 11, 12, 13, 15, 29, 38, 46, 50,

53, 70

safety, in nontoxic fumigant use, 38-40

Salmdorf Pietà, fumigation of, 77f, 77-78

San Francisco Maritime National Park, 49

Santoro, E., with Koestler, xi

Schnee-Morehead, Inc., 26

sculpture, and argon, 15t

Seal King, 24, 25t

Sealector II, 24, 25t

Segal, M., with Cipera, 31, 68, 75

sensors, temperature, 37

Shield Pack Class A, 19t

SM 5144, 26

Smith, C. P., 46, 68, 70


Smithsonian Institution, 68, 73-74

Snetselaar, R., 43

Society for the Preservation of New England

Antiquities (SPNEA), 70, 71-72

Index

spade-tip tacking iron, 24, 25t

Spanish piano, treatment of, 54f, 54-55

spatula tacking iron, 24, 25t

spiracles, 1f, 1-2, 2f

SPNEA. See Society for the Preservation of

New England Antiquities

storage cases, nitrogen-filled, 8-9

Stored Grain Research Laboratory, 8, 22

Story, K., 7

Strang, T., 13, 43, 74


suppliers, contact information, 81-89

Swagelok portals, 26, 27f

Ttacking irons, 24, 25t

spade-tip, 24, 25t

spatula, 24, 25t

Teledyne Analytical Instruments, 34

Teledyne trace-oxygen analyzer, 10, 14, 44, 59

temperature, 2. See also thermal techniques

and kill time, 5f, 11

temperature control systems, suppliers of, 88

temperature indicator paints, 37

temperature indicator strips, 37

temperature monitors, suppliers of, 88-89

temperature sensors, 37

tent, for anoxia, 57f, 57-59, 58f

tetrachloroethylene, 60

textiles, and argon, 15t

thermal techniques, 13-14

thermistors, 37

Thermo Lignum Company, 14

thermocouples, 37

thermometers, 37

Thomson, R. S., 13

three-jar humidification system, 29f, 29-30,

30f

Thylodrias contractus, 68

treatment of, 60

Tineidae, 9

trace-oxygen analyzers, 10, 14, 44, 46, 59

treatment containers, suppliers of, 89

Tribolium, and nitrogen anoxia, 9

tropical warehouse moth, kill times for, 69t

turnkey unit, 66

Unger, A., 50

with Reichmuth, 49

105

Unger, W.

with Reichmuth, 49

with Unger, 50

Union Carbide Company, 20

United States, carbon dioxide fumigation in,

70-74

University of California, Riverside, 53

VVacudyne chamber, 29f, 44

Vacudyne Company, 65-66

Vacufume chambers, 65-66

Vacufume Fumigator-36, 65

vacuum chamber, in static-dynamic method,

17-18

Vaisala, Inc., 37

Valentín, N., xi, 2, 3, 4, 9, 11, 12, 13, 15, 17,

31, 32, 33, 38, 42, 44-45, 50, 54-55, 67


Valtronics, 36

vapor-barrier laminates, 18

varied carpet beetles

kill times for, 12t, 45-46, 69t

treatment of, 60

vehicles, treatment with carbon dioxide

fumigation, 74-75

vinyl chloride, 19t

vinylidene dichloride, 19t

WWAAC (Western Association of Art

Conservation) Newsletter, 42

Warren, S., 74


webbing clothes moth, 8, 24

kill times for, 11, 12t, 45-46, 72

treatment of, 60

weight loss, 2-3

western drywood termites

burrowed, 15

kill times for, 12t

Whistler, Rex, 45

Wigglesworth, V. B., 1

Winterthur Museum, Delaware, 70, 71

wood, 14-15

and humidity, 31

Zebe, E., 5

zeolite, 22

zip-locks, 25

zippers, 25

S

U

Z

106Index

107

Charles Selwitz received his Ph.D. in organic chemistry at the University of Cin-cinnati in 1953. After graduation he joined the Gulf Research and DevelopmentCompany in Harmarville, Pennsylvania, where he was director of syntheticchemistry until 1982. He is now an independent consultant but works primarilyfor the Getty Conservation Institute in such diverse areas as the preservation ofearthen architecture by backfilling, the use of polymers for stone preservation,the stabilization of historic adobe with chemical consolidants, and the control ofinsect pests in museums by use of modified atmospheres.

Shin Maekawa is senior scientist in the Scientific Program at the Getty Conserva-tion Institute, where he coordinates nitrogen anoxia projects. His research areasinclude nitrogen environments for long-term storage and display of artifacts,disinfestation of museum objects, performance evaluation and design of storageand display cases, and monitoring and control of the environments and microen-vironments of artifacts. He is a registered professional engineer (P.E. Mechanical)in the state of California.

About the Authors

ISBN

9 780892 365029 90000

inert gases in the control of museum insect pests (1998)

Documents